{ "archive_path": "archive/1788689050.032601", "base_url": "lexfridman.com/dhh-2-transcript", "basename": "", "bookmarked_date": "2026-09-06 10:04", "canonical": { "archive_org_path": "https://web.archive.org/web/lexfridman.com/dhh-2-transcript", "dom_path": "output.html", "favicon_path": "favicon.ico", "git_path": "git/", "google_favicon_path": "https://www.google.com/s2/favicons?domain=lexfridman.com", "headers_path": "headers.json", "htmltotext_path": "htmltotext.txt", "index_path": "index.html", "media_path": "media/", "mercury_path": "mercury/content.html", "pdf_path": "output.pdf", "readability_path": "readability/content.html", "screenshot_path": "screenshot.png", "singlefile_path": "singlefile.html", "warc_path": "warc/", "wget_path": "lexfridman.com/dhh-2-transcript/index.html" }, "domain": "lexfridman.com", "extension": "", "hash": "YJYPHF2K4DBMMTDKX2C4", "history": { "archive_org": [], "dom": [], "favicon": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--max-time", "120", "--output", "favicon.ico", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://www.google.com/s2/favicons?domain=lexfridman.com" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-09-06T10:06:36.032775+00:00", "index_texts": null, "output": "favicon.ico", "pwd": "/data/archive/1788689050.032601", "schema": "ArchiveResult", "start_ts": "2026-09-06T10:06:35.573123+00:00", "status": "succeeded" } ], "git": [], "headers": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--head", "--max-time", "120", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://lexfridman.com/dhh-2-transcript/" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-09-06T10:06:36.504167+00:00", "index_texts": null, "output": "headers.json", "pwd": "/data/archive/1788689050.032601", "schema": "ArchiveResult", "start_ts": "2026-09-06T10:06:36.041082+00:00", "status": "succeeded" } ], "htmltotext": [], "media": [], "mercury": [ { "cmd": [ "/app/node_modules/@postlight/parser/cli.js", "https://lexfridman.com/dhh-2-transcript/" ], "cmd_version": "1.0.0", "end_ts": "2026-09-06T10:07:03.725110+00:00", "index_texts": null, "output": "mercury", "pwd": "/data/archive/1788689050.032601", "schema": "ArchiveResult", "start_ts": "2026-09-06T10:06:54.939116+00:00", "status": "succeeded" } ], "pdf": [], "readability": [ { "cmd": [ "/app/node_modules/readability-extractor/readability-extractor", "/tmp/tmp5dokhvf5", "https://lexfridman.com/dhh-2-transcript/" ], "cmd_version": "0.0.11", "end_ts": "2026-09-06T10:06:54.929109+00:00", "index_texts": [ "Episode highlight\n \n DHH \n (00:00:00) \n There are decades where nothing happens and weeks where decades happen. And we have seen decades of progress happen in the last nine months. If you\u2019re not recognizing the gravity of the moment, that\u2019s the delusion. That\u2019s the psychosis. Who would not get delirious if suddenly a genie pops out of the bottle and says, \u201cYou can have whatever you want. Every feature you\u2019ve ever dreamed of in an operating system, I can deliver them to you, most of them in five minutes, a few in 20, and if we really go hog wild, it\u2019s gonna take me two hours.\u201d I want the fastest car breaking the speed limits. I want the diver watch that can go down the Mariana Trench. I want the operating system that can install in less than 60 seconds.\n \n \n DHH \n (00:00:48) \n One of the things I always loved about race cars was when I would stumble out of the car absolutely smashed and barely able to hold my head up, and I\u2019d lay down on the garage floor and just think, \u201cHoly fuck, I\u2019m alive.\u201d The Overton window does not open itself. It opens one nudge at a time by people risking a little. A little reputation, a little pushback, a little criticism, or maybe sometimes a lot of reputation or a lot of criticism or a lot of pushback. Creating life with another human that you love is literally the peak experience of being on the planet.\n \n \n Introduction\n \n Lex Fridman \n (00:01:28) \n The following is a conversation with David Heinemeier Hansson, also known as DHH, creator of Ruby on Rails, CTO of 37signals, creator of the new Omarchy Linux operating system, best-selling author, race car driver, and one of the most outspoken and prolific programmers in the world. For the past 20-plus years, for him, programming meant meticulously handcrafting beautiful Ruby code. But recently, since late 2025, DHH has embraced the AI revolution and has openly transformed himself into, once again, one of the most outspoken and prolific agentic engineers, even though he hates that term. So let\u2019s say practitioners of whatever programming is becoming where AI is doing most of the actual programming and the human steers the ship with high-level design, vision, and taste. Yes, it\u2019s true.\n \n \n Lex Fridman \n (00:02:28) \n DHH is at times controversial, but he\u2019s always fearless, brilliant, and fun to talk to. This was, once again, an intense, eye-opening, and a wild rollercoaster ride of a conversation. This is a Lex Fridman podcast. To support it, please check out our sponsors in the description where you can also find links to contact me, ask questions, give feedback, and so on. And now, dear friends, here\u2019s DHH.\n \n \n Programming with AI agents\n \n Lex Fridman \n (00:02:56) \n 13 months ago, we sat down right here to talk about programming, and then everything changed. At that time, you were a bit skeptical about the role of AI in the process of programming, and then we went through this rapid evolution of agentic engineering. So simple question to start: How has your view on AI\u2019s role in programming changed? Are you excited? Are you terrified?\n \n \n Lex Fridman \n (00:03:23) \n Are you on an emotional rollercoaster ride?\n \n \n \n \n Lex Fridman \n (00:03:29) \n You\u2019ve been talking about fun a lot.\n \n \n DHH \n (00:03:31) \n There is none of the existential threat. That doesn\u2019t exist for me as an emotional component. It is only there as an intellectual component, and the emotional component for me is 100% pure, unadulterated joy- \u2026 and optimism and amazement that we\u2019ve made computers do this. And I find it so interesting that we talked just 13 months ago because it\u2019s like we talked in different universes, different eras. I love this quote. I think it\u2019s Lenin. There are decades where nothing happens and weeks where decades happen. And we have seen decades of progress happen in the last nine months. I mean, imagine you\u2019re there when the Wright brothers take flight.\n \n \n DHH \n (00:04:21) \n Yesterday, the New York Times would write, \u201cIt\u2019s gonna be 10,000 years before we fly,\u201d and then the day after, we\u2019re up in the skies, and just a few years after that, there were cross-Atlantic planes. The whole world has completely changed. What a blessing to be there in that moment. If you zoom out and look at all of human history, how many humans got to live within the same epoch that they were born in? They never saw that complete change of the world and of society. And to have been blessed with two of those feels just such a privilege. I got to see the internet- \u2026 from pre-internet to post-internet, and then now pre-AI, post-AI. What an amazing run. How fortunate.\n \n \n Lex Fridman \n (00:05:12) \n Yeah, but the thing is, this feels like a thing that happened faster than anything else in human history.\n \n \n \n \n Lex Fridman \n (00:05:18) \n So I would say somewhere around December, maybe late November-\n \n \n \n \n \n \n \n \n Lex Fridman \n (00:05:26) \n You know, when they talk about when one nation invades another in a world war or something like this, it\u2013 this is how we talk about AI changing everything. Yeah, the\u2026 And it really just shifted for many great developers. It shifted to where AI is doing some basic autocomplete, maybe writing 5, 10, 15, 20% of code, to writing 80% of code.\n \n \n \n \n Lex Fridman \n (00:05:50) \n Or 100, yeah, especially if it\u2019s not a public-facing product.\n \n \n DHH \n (00:05:54) \n To me, that\u2019s why even when I look back upon our conversation a year ago, I don\u2019t actually have different opinions. I have the same opinions. A year ago, I did not like the mode of AI we were offered. It was the autocomplete mode, or it was the AI chatbot mode. Now, the chatbot I actually liked, as we talked about. Great tutor right from the get-go, great way of looking things up on the internet, not what was gonna replace me chiseling code. But- \u2026 then we get the agents.\n \n \n \n \n DHH \n (00:06:28) \n And the agents start out being curiosities for about five minutes, and then they get amazing, and then they get, \u201cOh my God, is this AGI?\u201d And all of that happened since last year, even just in, within the last nine months. We have basically these few phases here. We have AI in the pre-agentic era. I was excited about that, but it was not fundamentally rewriting the rules of the game for me. It was not completely changing how I worked. I was still chiseling code, I just had a little helper, a little sidekick-\n \n \n DHH \n (00:07:08) \n \u2026 who could bounce ideas off, and I could look up this information online and so forth in a more efficient way. It was just a more efficient way to do what I was already doing, and it didn\u2019t change the emotional connection I had to the computer. Then we get to November 24th, 2025. Opus 4.5, to me, was the dividing line, where suddenly\u2026 I didn\u2019t even try it on the 24th. I think I tried it on the 26th. I gave it a couple of tasks, and I realize that the quality of the output is uncannily close to what I would\u2019ve written. And I remember just leaning back and thinking, \u201cWhat just happened?\u201d\n \n \n DHH \n (00:07:55) \n \u201cHow did we go from this autocomplete mess that I was talking to you about in the summer to this just a few short months later? How did we get both the increase in intelligence and then also the increase in usability?\u201d This agent harness question where I don\u2019t know if Opus 4.5 was that much smarter than Opus 4, which is what we had in the summer, but its ability to instrument your computer, to use tools, to check its own work, to apply its intelligence in such a way that you could get real meaningful work out of it, was completely different. And I think this is then the big change that happens for almost anyone who paid attention and started playing with it over the Christmas break. This is what I heard-\n \n \n DHH \n (00:08:44) \n \u2026 from Shopify and other places with lots of employees who suddenly had a breather, suddenly had a couple of weeks to lean back and just look at what was going on, gave it a try, and had the same mind-blowing experience that these agents were of a different genre than what we had before. And then what we get to is, I\u2019m already excited at this point. Like, by December, I\u2019m already revisiting sort of all my priors- \u2026 and going like, \u201cWow, if it can do this, can it also do that?\u201d \u201cOh, yeah, it can.\u201d And again, Opus 4.5 now looks like a retarded model. And this is the magic of this progress, is you think you\u2019ve reached something\u2026 I remember thinking at the moment, \u201cIf this is the last model we get, I\u2019ll be set. I\u2019ll be happy.\u201d\n \n \n \n \n DHH \n (00:09:39) \n I could live with Opus 4.5 for the next 20 years- \u2026 and you would hear no complaints from me because it was just so incredibly amazing to see an agent do all this work in the way that I wanted it done. Because it was not just about it being able to solve a task, it was also that I could look at its path there and go, \u201cYep. Yep. Maybe not there, but almost,\u201d and it would give you two notes. You\u2019ll get to where I wanted to go in the way I wanted to go there. You could produce code I wanted to merge. You could produce code that actually looked beautiful if it was written in Ruby. Rust, different question. But this ability for the agents to truly become an extension of how I wanted to work was very novel. But then we wait just until early spring, and suddenly we get sub-agents.\n \n \n DHH \n (00:10:37) \n We get harnesses that can subdivide the task, and something that would take Opus quite a while suddenly took a fifth of the time, a 10th of the time, because it could get chopped up and suddenly you got eight sub-agents working for you. But both of those two first phases of the agentic age, to me, still felt like I had to drive. I could tell it what I wanted, where to look for it, and steer it a little bit when it went off, and then we\u2019ll get there, and I would go much faster.\n \n \n DHH \n (00:11:11) \n But I had to be in the driver\u2019s seat. I had to tell it what I wanted, and I had to be the reviewer, the auditor of what was coming out. And then finally now, this summer, with Opus 5, Fable, and Sol, GPT Sol, and to a lesser extent, some of the open weight models, we\u2019ve arrived at a new era where I\u2019m not telling it where we\u2019re going. I\u2019m telling it the problem I have. I\u2019m telling it the fuzzy, vague idea I have. It tells me where we\u2019re going. It tells me which path to take, and I will still look at it because I\u2019m a curious person and I like computers and I like the outcome of it, but I really kinda don\u2019t have to. I\u2019ve become optional in the part that produces the code that picks the route. Remember when early GPS systems came out?\n \n \n DHH \n (00:12:09) \n They were amazing compared to looking at a map, but you still wanted to pay attention. Is it gonna drive you in the harbor? I remember these newspaper articles. \u201cAh, GPS, they\u2019re terrible because people don\u2019t pay attention, they drive in the harbor.\u201d When was the last time GPS drove anyone in the harbor? Like, that just doesn\u2019t happen anymore. In fact, the cars now just drive themselves, right? And this is where we\u2019ve arrived at now, that I can trust. For the domain I\u2019m working in right now, I can trust it and feel completely confident that it\u2019s gonna have my back. It\u2019s not gonna do something stupid, and if it does something stupid, it\u2019s gonna be able to recover.\n \n \n Lex Fridman \n (00:12:45) \n Mm-hmm. We should mention that there\u2019s all kinds of domains that-\n \n \n \n \n Lex Fridman \n (00:12:48) \n \u2026 programmers operate in. There\u2019s\u2026 I actually don\u2019t know, but I think the most common domain in development is, like, web dev CRUD. You have a database.\n \n \n Lex Fridman \n (00:13:00) \n You have a user-facing interface, and it does something back and forth, and it could be internal to just one person, to multiple people, to a small number of people, or to the world. And I think for that, it\u2019s\u2013 I mean, you could really get to the 100% of code written by AI and really almost\u2013 if you\u2019re a good programmer and you have a good intuition about what\u2019s happening behind the scenes, you can legitimately not look at the code, at least that\u2019s been my experience. I don\u2019t know how much prior experience with programming you need to have to kinda know that things are working correctly behind the scenes, like intuitively by observing the ripple effects, the symptoms of the system.\n \n \n Lex Fridman \n (00:13:45) \n But I think for that domain it\u2019s close to 100%, and then there\u2019s domains like you\u2019re also operating in, which is writing a Linux distribution. There, maybe there\u2019s more because you\u2019re obsessed with speed and all that kind of stuff. There, maybe you need to look at the code a little bit more. Then there\u2019s maybe safety critical systems all the way down to operating a nuclear power plant or a self-driving car. Maybe you need to look at the code more carefully.\n \n \n DHH \n (00:14:12) \n Yes, but that said, AI is insanely capable at both finding-\n \n \n \n \n DHH \n (00:14:17) \n \u2026 and fixing security vulnerabilities. This was the whole blowup about Fable. This model was so capable of finding holes that an attacker could exploit that it was simply not safe to release. So the irony here is that when you look at that field, it seems like we\u2019ve reached levels of intelligence that virtually no human can match because many of these security holes are about stringing combo moves together. You find one little vulnerability here that by itself might not be the worst thing in the world, but then you combine it with four others, and suddenly you have RCE, remote command execution. Humans who are able to do that are very rare. They usually work inside state-sponsored organizations or other clandestine operations. They\u2019re not just out and about finding things.\n \n \n DHH \n (00:15:14) \n So they\u2019ve gotten so good at that. The Linux distribution is why I\u2019ve gotten 100% AI pilled. Because I\u2019ve been working on Omarchy for the last three months, this version that just dropped a few days ago called Quattro. And almost right from the beginning I\u2019m working on that version, the agent acceleration neared 100%, and in the last two months it has been 100%. I have not written-\n \n \n \n \n DHH \n (00:15:43) \n \u2026 any of the code that\u2019s shipped in Quattro by hand. I\u2019ve reviewed the shape of all of it. I\u2019ve reviewed the individual lines of anything that\u2019s critical in the model layer of the system, and I\u2019ve not looked at a bunch of the UI code. I have not looked at a bunch of the auxiliary code, and I\u2019ve not written any of the new functionality entirely by hand. But then the web part, actually evolving Basecamp and Hey, our professional products that have lots of users and are relatively large code bases, have proven surprisingly tricky to fully accelerate with agents. We just released Basecamp 5 not too long ago. That was the first product at 37signals that was really agent accelerated. Because we were in this final sprint phase from around February. By then, agents were already good.\n \n \n DHH \n (00:16:44) \n And we had this early surge of, \u201cIt\u2019s solved. We can just have the designers do the programming. They know what features they want. They know what shape they want it to take. Let them vibe.\u201d And we let them vibe. And we ended up with a lot of PRs that individually perhaps could have been justified for a hot moment, but taken all together, destroyed the architecture of the system. And we actually had to clean up manually, mop it up by hand, by human hand, to get back to an architecture that felt cohesive and coherent. So we still have a bit of that\u2013 that was February, by the way. Things are quite different now.\n \n \n Lex Fridman \n (00:17:25) \n Well, hold on a second. What\u2019s the lesson from that? Is one of the lessons from that that you have to be a programmer at this stage to be able to vibe code?\n \n \n DHH \n (00:17:33) \n To be able to vibe code on existing substantial-\n \n \n \n \n DHH \n (00:17:38) \n \u2026 code bases, even if they\u2019re crud, if you wanna retain the element of architecture that got that system to where it was. Now, that\u2019s also a point where I\u2019ve stressed many times that when people accuse vibe coders of being slob generators, I go right back at them and say, \u201cHave you looked at the average programmer\u2019s output?\u201d That is some other slob too. If you\u2019ve looked at behind the scenes of many great companies and what their code bases look like after there\u2019s been 3,000 humans through them, they\u2019re awful. Absolutely awful.\n \n \n How software will change\n \n Lex Fridman \n (00:18:14) \n Can you tell me the intuition you have? I seem to\u2014just stepping back and observing the different apps that we all rely on, I don\u2019t know, Adobe Photoshop, all this kind of stuff\u2014there seems to be\u2026 the progress on development there has not accelerated. So why is it, what lessons can you draw from Basecamp, like well-established, huge user base? Why aren\u2019t we seeing, like, super rapid increase in, like, new updates, features, all this kind of stuff in these well-established apps?\n \n \n DHH \n (00:18:44) \n Multiple reasons. I\u2019ll start with the first one that\u2019s the most critical. As soon as you\u2019re having human teams work together on something, the bottleneck is rarely implementation. It\u2019s human bandwidth and communication. When you have a product manager and a couple of designers and a VP above them and a CTO above them, and everyone wants to be part of the shaping process because we\u2019re all justifying why we\u2019re here, that\u2019s where all the productivity goes to die. The revelation I\u2019ve had working on Omarchy the last three months is that to get that magical 10X, 100X, in a few rare cases, 1000X productivity boost, you have to interact with the agents directly, and you cannot intermediate that bandwidth with another human because it\u2019s simply too slow.\n \n \n DHH \n (00:19:49) \n And on the one hand, that\u2019s a bit of a bummer. I mean, I like humans, and it\u2019s great to work together. But it also means we need to temper our expectations with what these agents can do if it\u2019s humans driving it, and you have three layers of approval and all the other machinery of a large corporation. The implementation part is only a small segment of it. The other thing I\u2019d say is that most organizations don\u2019t know what they want. They don\u2019t know how to make it better. They\u2019re not bottlenecked on implementation. They\u2019re bottlenecked on ideas. They\u2019re bottlenecked on vision. They\u2019re bottlenecked on taste. And if you don\u2019t have those elements in excess of your implementational capacity, it doesn\u2019t help. So you can make a lot of shitty ideas come true, then what? Are you gonna ship that?\n \n \n DHH \n (00:20:46) \n That sounds like, I don\u2019t know, something coming out of Microsoft. That\u2019s not what we\u2019re trying to replicate here. That process is not actually\u2026 Because we already had this. If you step back for a moment and think of the towering organizations we have who have had tens of thousands of programmers at their disposal. I mean, I\u2019m picking on Microsoft here. I love Microsoft some of the time, but I\u2019ll pick on them in this case because they have had endless resources, endless programming capacity for decades, right? This is what\u2019s showing us that just being able to write a lot of code does not produce great compelling software. Now, the other thing is we\u2019ve had this capacity for about six months. That\u2019s not very long in human life cycle capacity of internalizing what\u2019s going on.\n \n \n DHH \n (00:21:40) \n And I think it\u2019s actually funny that the critique of AI is, why isn\u2019t it going faster? What are you talking about? We\u2019ve not had any other form of progress that has moved as fast as AI, and you\u2019re impatient because in the last three months we haven\u2019t rewritten the whole world and made it a utopia of software goodness?\n \n \n Lex Fridman \n (00:21:59) \n You\u2019re basically saying everybody should be switching to Linux now. One of the arguments is, like, we can rewrite all the software that\u2019s not available in Linux in Linux. Linux has been the love of my life for many years, but one of the reasons I\u2019m still attached to Windows and now Mac is because of video editing, Premiere.\n \n \n DHH \n (00:22:15) \n We\u2019re gonna fix that. But-\n \n \n Lex Fridman \n (00:22:17) \n So the question is, who\u2019s gonna build Premiere and Photoshop for Linux? And it feels like one person can now.\n \n \n \n \n Lex Fridman \n (00:22:28) \n And so, and I\u2019m allowed to be impatient. In fact, me or anyone else, impatience is the first step of doing it yourself, right? Like-\n \n \n \n \n Lex Fridman \n (00:22:40) \n \u2026 and there\u2019s a community of tens of thousands of people that use Adobe Premiere or DaVinci, all these non-linear video editors, that know the frustrations. Everybody shares them. You can look at Reddit. You can look at the forums. Everybody knows it. I don\u2019t know what it is. It may be the management. It may be the meetings. All the things you\u2019ve always discussed is the bureaucracy that\u2019s inside companies that\u2019s slowing everything down in the development of features. But it feels like there\u2013 if you just let some of the developers loose with a bunch of agents, you can fix all that.\n \n \n Lex Fridman \n (00:23:13) \n But again, based on everything you\u2019re saying, maybe the way to do that is to start over from scratch, maybe through open source, or somebody launches a new company, that within large companies now it\u2019s hard to really accelerate into the agentic era by letting developers just build.\n \n \n DHH \n (00:23:33) \n This is the classic innovator\u2019s dilemma. These companies have gotten so good, so established at the old way, and therefore their entire structure, management layers, processes are tuned for a time that no longer exists. But you can\u2019t pivot that. These are super tankers. It just doesn\u2019t happen. This is why we\u2019re finally getting this upset in the technology industry. For a while, I was very upset about the duopoly between Apple and Google on mobile, because it felt like mobile was the most important computing platform that we had, and I could not see a path to unseat either Google or Apple sitting on top of that, controlling everything, taking their toll booth money. But the game has changed. This is no longer the most important platform.\n \n \n DHH \n (00:24:32) \n Mobile phones are important, but there\u2019s also a lot of other form factors coming, whether it\u2019s glasses or it\u2019s earpieces or whatever else have you. It\u2019s all in play now, but the computing platforms themselves are in play for the first time in probably 40 years, if you look at the desktop. Linux has been around since \u201991. It\u2019s not taken off or taken over on the desktop. It\u2019s taken over everything else. All the devices you have on your desk, your fridge, your toaster, everything runs Linux except for your computer. Now, funnily enough, your Android is actually Linux, but wrapped so sufficiently that you can\u2019t really recognize it. But now there\u2019s an opening, and the opening is exactly as you say it.\n \n \n DHH \n (00:25:21) \n If there\u2019s a piece of software you depended on that bound you to Windows, it is completely within reach for you personally to start rewriting it. And maybe you won\u2019t get to 100% coverage, but this is the old joke about Microsoft Office. \u201cI only use 5%.\u201d Yeah, well, we all use a different 5%. Well, what if we all just build our own 5%? What if I just took the functionality that I need and just did that? That is a completely different challenge and one agents are incredibly capable of doing right now today, and I\u2019ve done it a lot of times over. So one of the amazing things I\u2019ve found with the latest agent engaged is I have become a polyglot programmer, something I was absolutely not before. I was a Ruby programmer- \u2026 first and foremost, and then I dabbled a little bit in Bash-\n \n \n DHH \n (00:26:17) \n \u2026 when I had to. And in the last two months, I\u2019ve written C++. I\u2019ve written three applications that have shipped in Omarchy Quattro. I wrote a writing app. I was using this app called Typora, which is a very nice app-\n \n \n \n \n DHH \n (00:26:33) \n \u2026 which in itself is based on another app called iA Writer, which was my real love on the Mac- \u2026 for a clean, simple Markdown writing environment. That\u2019s where I write all my essays.\n \n \n DHH \n (00:26:45) \n And then I moved to Linux, and I couldn\u2019t get iA Writer, so I moved over to this other tool called Typora. It was a piece of shareware, and it had a lot of features I just didn\u2019t need. And about six weeks, seven weeks ago, I thought, \u201cDo you know what? I only need 5% of Typora,\u201d which is already a basic app. I literally told the agent to get going, that I wanted it written in C++ and Qt, because that would fit well with the aesthetic of what I was building with Quattro. And in, I think, about 20 minutes, it had the first version. I started using it, and it wasn\u2019t quite right. Within two days, I\u2019d given up Typora, and I wrote and have written all of my essays since that moment in Omawrite.\n \n \n AI impact on open source\n \n Lex Fridman \n (00:27:30) \n Can I ask for your advice and your vision about something? So what I find with agents, you can actually write something like a replacement for Typora that is perfectly customized to you, to your needs- \u2026 with just one user in mind. And then there\u2019s a version of that that starts with one user in mind but expands to a larger audience. So I\u2019ve actually written a large number of software just for me, and I find it easier to do and better to get my situation done. For example, I have a video editor I\u2019ve written, which is very easy to do just for me.\n \n \n Lex Fridman \n (00:28:06) \n I\u2019ve never, unlike you, built something that\u2019s used by a large number of people, and I find that a bit scary and intimidating. So, like, at this, with this agentic age, how do you take a step to actually build something that solves your problem, which I think is a beautiful way to build, but also expand it to actually useful for other people? Like, is there advice you can give, and do you also see the problem of how easy it is to just build a tool for yourself?\n \n \n DHH \n (00:28:35) \n Yes, but it is simpler than you think. By the time you\u2019re done building the tool for yourself, you simply tell your agent to put it on GitHub. You don\u2019t even have to do anything else. It will figure out how to put that repo on GitHub. It\u2019ll write a nice README. It\u2019ll start using GitHub releases so that you can track things. It\u2019ll actually be a better software maintainer than you could ever be- \u2026 because it is far more patient, it is far more diligent in doing the drudgery of managing open source software than you are.\n \n \n DHH \n (00:29:05) \n And you can simply get on with the joy of using the software that you built and evolving it as you see fit. Now, this gets into the big argument we\u2019ve been having in open source for a while. Are AI contributions good or bad for the maintainer? There\u2019s a lot of maintainers right now who are quite upset about the fact that they\u2019re suddenly getting a huge influx of pull requests, of contributions from people who may not be the best programmers or programmers at all- \u2026 towards their software. I look at that argument and go, \u201cAre you kidding me?\u201d Here is a vein of free contributions that you can take or don\u2019t take, but you\u2019re complaining about the fact that they\u2019re there? It sounds straight out of- My steak is too juicy and my lobster too buttery. What are you complaining about?\n \n \n DHH \n (00:30:07) \n This is the most amazing thing. This is what we proclamated for so long was going to be the magic of open source, that we could all contribute to it, but it was, of course, never true. There was only a small subset of people who contribute to open source, and that was the very highly skilled wizards. So this is a bit of a reformation moment. We had this disintermediation between us and the computer, and it was this class of clerics and priests called programmers, and suddenly they\u2019re being disintermediated here by was it the 95 theses by Luther in 1500s, nailing them to the door? And one of those was very clearly about the intermediation, that you should have direct access to the higher intelligence.\n \n \n Lex Fridman \n (00:30:51) \n But don\u2019t you need the wizards to keep a high bar of excellence in the code base?\n \n \n DHH \n (00:30:56) \n 100% you do, which is also something that\u2019s always been true about open source, which is the second argument that really grinds me. I\u2019ve been running open source projects for 25 years. I have literally looked at the output of thousands, if not tens of thousands of programmers. I feel like I have the statistical basis to assert that most programmers, they suck. And I don\u2019t mean that in the way it sounds.\n \n \n DHH \n (00:31:24) \n That was why I paused for a hot second here. I mean, they suck in the sense that they don\u2019t write the code I want to have written. They don\u2019t prepare their bug reports with all the relevant information. They don\u2019t detail their pull requests with the why. They don\u2019t bother to fill in needed code comments. They don\u2019t double-check their work. They don\u2019t write unit tests. They don\u2019t do all of these things it takes to create good software. But do you know who\u2019ll do all that stuff? Agents, if you tell them to. They\u2019re very diligent at following your instructions, most of the time, sometimes to a fault. But it actually means that if you take the median programmer and their pull requests towards an average open source project, they\u2019re already getting outclassed by agents.\n \n \n DHH \n (00:32:11) \n I would rather get an agent-written pull request to one of my projects than I\u2019d get one written by a human, and it\u2019s not just because the quality\u2019s better. It\u2019s also because I feel a lot less bad if I just reject it. You didn\u2019t even write it, so I can simply look at what your agent wrote on your behalf and go, \u201cEh, don\u2019t want it.\u201d That\u2019s always been true in open source, and in fact, I think this is one of the problems with open source maintainers. They have way too much ball of neuroticism and anxiety that requires them to look upon any contribution as an obligation on their part to do everything to get that to land in the code base. That\u2019s not true at all.\n \n \n DHH \n (00:33:00) \n You can simply say, \u201cNo,\u201d or even better, you can say, \u201cNo thank you.\u201d And if you get into the mode of realizing that your project is allowed to exist and evolve according to your vision, according to your roadmap, it becomes so much easier in the agentic age to decline the contributions you don\u2019t want because you don\u2019t even inconvenience or hurt the feelings of a human. It\u2019s just a clanker, and the clanker won\u2019t mind. In fact, the producers of clankers, the labs, love when you spend tokens in vain. They get paid just the same. So, in my opinion, this is the absolute best time to ever have been an open source software maintainer.\n \n \n DHH \n (00:33:46) \n Not only do we get this wealth of glorious pull requests made by agents with all the boxes ticked, we also get to tap into a creativity of people who did not have access to contribute to a project before. I\u2019ve been running the Omarchy project now for a little over a year, and in the last three months working on Quattro, I have merged over 1,000 pull requests. Quite a lot of those pull requests were written by people who were not classical programmers, or were programmers in other domains, not Linux operating system or distribution development. They were able to contribute their good ideas because of agents, and I got to cherry-pick the very best of that-\n \n \n DHH \n (00:34:32) \n \u2026 because it was there. Otherwise, those ideas would just have lived inside their heads. Again, is that not the purpose of open source, that we tap into the collective intelligence and creativity of the whole goddamn planet, and we channel that towards a commons where we all benefit from the combined efforts of everyone? Now, there\u2019s currently, I think, about 400 unmerged pull requests on Omarchy, about double what it was a week ago. So the acceleration is real, but so are the tools. I\u2019m not reviewing every pull request anymore. I haven\u2019t been reviewing them for quite some time now. I have the agents- \u2026 review them for me- \u2026 and then they will give me a summary about whether something is ready for the decision, the human decision. Should we merge or should we not merge?\n \n \n DHH \n (00:35:25) \n I don\u2019t need to look at all the pull requests that are either wrong, duplicated, bad. An agent can sort that chaff away from me. I just get to look at the pearls- \u2026 the good ideas that are ready to go, the bug fixes that the agent has validated in a VM on my behalf. All the drudgery of replicating and managing open source projects is evaporating at lightning speed, and we are left with the golden juicy parts, the bone marrow of software development- \u2026 deciding what should this thing do and where should it go?\n \n \n Lex Fridman \n (00:36:03) \n Do you think the good ideas, ultimately the kernel of the good ideas originated in the human mind, so the individual contributors Like, can it be done by agents, or is the PR ultimately requires a human idea? Like how to improve Omarchy, all those incoming\u2026 Or can it all just be a pool of agents?\n \n \n DHH \n (00:36:24) \n That was the mode and the way I was thinking about agents in the first agentic age from November 24 to February 28. I thought all the ideas would originate with the humans. They would tell their agents what to build, and off they went. I don\u2019t think that\u2019s true anymore at all. I have seen things you people wouldn\u2019t believe, ideas coming out of models so great that it makes me humble- \u2026 as a person who otherwise prides himself on having good ideas. The agents are incredibly capable of creative thought, and folks who are still stuck in the analysis that agents are parrots- \u2026 just regurgitating the ideas that are already there, are delusional about the progress that\u2019s been made in the last six to nine months.\n \n \n Building Omarchy Linux distro\n \n Lex Fridman \n (00:37:21) \n So some people listening to you right now will say DHH is suffering from the old case of AI psychosis. Can you steel man the case that you are in fact experiencing in a state of delusion, like One Flew Over the Cuckoo\u2019s Nest, and can you argue against it?\n \n \n DHH \n (00:37:37) \n I\u2019m in a state of delirium. That\u2019s the state I\u2019m in. Because I have been working with computers for 40 years, and I\u2019ve never seen the things that I\u2019ve seen in just the last two months.\n \n \n DHH \n (00:37:54) \n A whole career, a whole life dedicating to the love of computers as my main focus for the working hours and then some, suddenly completely upended, suddenly completely rewritten. Of course I\u2019m delirious. Who looking at this reality is not? If you are not delirious or at the very least very excited\u2026 Actually, I guess that\u2019s not even true. If you\u2019re not recognizing the gravity of the moment, that\u2019s the delusion. That\u2019s the psychosis. The psychosis is believing that the world is barely different. We just have some electronic parrots reciting things to us. Now, in my opinion, the reason this argument is actually brought up is because people are not able to see the fruits of the progress. That goes to your point earlier. Where is the amazing software?\n \n \n DHH \n (00:38:49) \n Why isn\u2019t American GDP running at 12% year over year? First of all, give it a minute. It hasn\u2019t even been a year, goddammit. But second of all, I feel on my own account, I have actually brought the pudding. Omarchy Quattro launched on Friday. It\u2019s been downloaded by tens of thousands of people. And they like it. They like it a lot.\n \n \n Lex Fridman \n (00:39:17) \n We should mention that going to Perplexity here, Omarchy is a highly opinionated Arch Linux-based desktop distribution centered on the Hyprland Wayland tiling compositor. It was created by David Heinemeier Hansson, DHH, as a polished developer workstation set up with a cohesive visual style and many everyday tools already configured.\n \n \n DHH \n (00:39:39) \n Let me put it. Omarchy is a beautiful, modern, and opinionated Linux system. It\u2019s an alternative operating system to macOS and Windows, and it\u2019s freaking amazing.\n \n \n Lex Fridman \n (00:39:50) \n And the Quattro thing is you\u2019re pushing towards the age of agents.\n \n \n DHH \n (00:39:54) \n That\u2019s right. That\u2019s the latest version that we just dropped out, which is also funny. The Omarchy project is only a year old and a little bit. I started it last summer in between sessions at the 24 Hours of Le Mans-\n \n \n DHH \n (00:40:08) \n \u2026 where I had watched one too many YouTube videos on Linux Ricing and got bitten by that bug. Started working on this second iteration of my attempt at creating a better Linux operating system. The first iteration we talked about last time was called Omakub, was built on top of an existing system called Ubuntu, and it was fine. But the ambition I was able to pour into the new version, Omarchy, because I started seven layers deeper down the stack, was completely different. But it was still done in a pre-agentic world. I wrote all those bash scripts by hand in the beginning, and then I got to see the midway point where things changed over.\n \n \n DHH \n (00:40:52) \n I did a couple of versions of Omarchy that had partial agent acceleration, and then three months ago, it was full throttle, 100%, everything is written by agents but steered by me. And that just ended up being a very different experience and a very different operating system because I was suddenly granted a limitless ceiling on my ambition. I could look at any feature in Windows, on Mac, on other Linux systems and say, \u201cI want that,\u201d and the agents would deliver. So everything I wanted was suddenly within reach, which meant that I got a little delirious. Who would not get delirious if suddenly a genie pops out of the bottle and says, \u201cYou can have whatever you want. Every feature you\u2019ve ever dreamed of in an operating system-\n \n \n DHH \n (00:41:47) \n \u2026 I can deliver them to you, most of them in five minutes, a few in 20, and if we really go hog wild, it\u2019s gonna take me two hours.\u201d\n \n \n Lex Fridman \n (00:41:54) \n Mm-hmm. Have you seen Requiem for a Dream, the movie? So there\u2019s a drug-like effect here. I find it personally\u2013 We\u2019ll talk about many levels of this, but it is overwhelming how much power is placed in our hands so suddenly, and it\u2019s very hard to know, it\u2019s like with \u201cThe Hobbit\u201d or something, what to do with it, and I find myself truly overwhelmed with the multitasking on the verge of almost burnt out. Super excited about so many things, but ultimately looking back, you want to focus on one thing and really ship.\n \n \n DHH \n (00:42:31) \n I don\u2019t have that sense. I don\u2019t have the sense of overwhelm. I don\u2019t have the sense of dread. I don\u2019t have the sense of distortion because I have a mission. I\u2019m going somewhere, and therefore, I can channel all this new power towards-\n \n \n Lex Fridman \n (00:42:48) \n Yeah, it\u2019s wonderful.\n \n \n DHH \n (00:42:48) \n a singular goal and outcome, creating the perfect computer. And therefore, all my investigations with AI are focused towards that end. Now, I also have a day job. And we also use agents there, and that\u2019s also targeted towards sort of specific outcomes. But with the Omarchy experience, I got to tap in straight in the back of my head and increase the bandwidth between ideas arriving in my brain and software emerging on the screen. It was like going from dial-up- \u2026 to fiber. I think on your podcast, Elon talked about the actual human bandwidth-\n \n \n DHH \n (00:43:35) \n \u2026 that Neuralink and other projects are trying to accelerate. Like, the bandwidth we\u2019re communicating over right now is quite low because we\u2019re limited by our cognition. We\u2019re limited by the rate of speech. You\u2019re not limited by those things when you\u2019re working with a swarm of agents. Now, let me caveat this by saying, if I\u2019d heard myself talk like this-\n \n \n DHH \n (00:44:03) \n nine months ago, I would probably have used the label AI psychosis. I think it would be fitting because nine months ago when people were talking like this, they weren\u2019t shipping, and I think that\u2019s the ultimate difference, that there were people who saw this early. I was not actually that early. As we\u2019ve talked about, I was rather skeptical, and I was using AI in certain ways, but I was not the first person who downloaded Claude Code. I think Boris released that in end of February. I don\u2019t think I installed Claude Code until September. So there was about six months there where pioneers saw these glimmers of what the future was gonna look like. It wasn\u2019t there yet because the intelligence wasn\u2019t there. The harnesses weren\u2019t there, but they could see the glimmers. I couldn\u2019t.\n \n \n DHH \n (00:44:48) \n Tobi, Tobi Lutke, the CEO of Shopify and now my partner in crime on this Omarchy project in part, he\u2019s gotten pilled with Omarchy as well. He saw these things very early, and he tried to tell me, and I wasn\u2019t seeing it. I was not seeing what he was seeing.\n \n \n \n \n DHH \n (00:45:08) \n Yes. I was still on Earth. And he was\u2026 He had already boarded the rocket and was heading towards space. I remember actually reading his memo to the company, the AI memo, and thinking, \u201cAh, seems a little much.\u201d Like, you\u2019re making a big deal out of something that I cannot yet feel is a tangible thing.\n \n \n DHH \n (00:45:32) \n Because where\u2019s the output of this? I try using it. It writes code I don\u2019t like. It wants to interrupt me all the time. Where are you getting this from? So I get it. I totally get it. If you\u2019ve not seen what the current quality of intelligence can produce, of course you\u2019re gonna look at someone and go like, \u201cThey sound a little nutty,\u201d because all your experience up until this point would tell you that they are nutty. Because if you look at the history of computer programming, for the last 40, 50 years, people have been promising that regular people could write code just by talking. We\u2019re gonna have fourth generation languages.\n \n \n DHH \n (00:46:12) \n Lisp was once upon a time presented, Smalltalk was presented as these environments that would let regular people write their own applications, and none of it came true in the sense beyond, like, Microsoft Access databases or Excel spreadsheets. Those are end user programming environments, but it\u2019s nothing like what we\u2019re able to do now. And then, I mean, I\u2019m not even talking about end user. I\u2019m talking as me, someone who\u2019s been programming for 25 years, seeing the output and the acceleration and the quality and then shipping it, and I think that\u2019s what\u2019s gonna change the entire conversation quite quickly.\n \n \n DHH \n (00:46:49) \n The few laggards who are still on the fence about whether this is actually useful, who are still trapped in a meme from early \u201925 about mechanical parrots, they\u2019re simply going to be overwhelmed by the evidence that\u2019s about to flood over them.\n \n \n Vibe coding vs agentic engineering\n \n Lex Fridman \n (00:47:05) \n So I hope to see projects like Omarchy, several of them that, like, really show agent first.\n \n \n DHH \n (00:47:15) \n It\u2019s coming. All of it is coming.\n \n \n Lex Fridman \n (00:47:17) \n In the spaces, in the silly spaces that I already mentioned, like video editing, all these kinds of apps that are like Typora- \u2026 this kind of stuff.\n \n \n DHH \n (00:47:26) \n It\u2019s gonna come, and I\u2019ve already seen the early inklings of this. One of the really neat features about Omarchy is that it has a far more robust plugin system that allows you to extend your operating system and rewrite it really in terms of its user interface and its panels and its features by creating your own plugins that can be cloned off of tools we ship in the box. For example, say you want a different calendar. Omarchy ships with a calendar. You click the little clock. It pops up a calendar, and it doesn\u2019t have iCal support, for example. It doesn\u2019t consume your appointments. There\u2019s about 17 implementations of that already, and in three days, we had 330 plugins on the Omarchy plugin marketplace. I have never seen growth like that with any project I\u2019ve ever been involved in.\n \n \n DHH \n (00:48:23) \n I\u2019ve never seen participation that broadly. I\u2019ve never seen that many people be able to create software that\u2019s meaningful for them\u2014also be usable for others so quickly. All of it is driven by the fact that Omarchy ships a set of skills that tells any agent you bring to it how to create extensions to the operating system, and therefore affording you the vision of the true malleable\u2026 I was about to say agentic. I hate that fucking word. And the reason in part I hate that word is, first of all, it\u2019s become marketing slop speak- \u2026 at this point. Like, it\u2019s just slapped onto everything. I wish we had a different word that just meant AI doing stuff.\n \n \n Lex Fridman \n (00:49:08) \n Vibe coding feels wrong, too.\n \n \n DHH \n (00:49:09) \n It does, because vibe coding to me smells exactly like script kiddies did in the early 2000s. People applying PHP scripts they just downloaded offline that they don\u2019t understand anything of. And while, I mean, that\u2019s true in the sense that much of the vibe coding, including my own vibe coded projects, and as I just mentioned, I have several. I\u2019ve written things in C++. I\u2019ve written things in Rust. Rust\u2014I hate Rust with a passion. Rust to me is like pouring acid in my eyes when I have to look at the code.\n \n \n Lex Fridman \n (00:49:42) \n Wait, wait, wait, wait. Why do you hate Rust?\n \n \n DHH \n (00:49:43) \n It is, in my opinion, the ugliest programming language-\n \n \n \n \n \n \n \n \n DHH \n (00:49:47) \n \u2026 in probably the last 40 years.\n \n \n Lex Fridman \n (00:49:49) \n So it\u2019s the opposite of Rails and Ruby.\n \n \n \n \n \n \n DHH \n (00:49:53) \n \u2026 I\u2019ve written applications in Rust because the output of Rust is actually amazing. The memory safety, the fact that it\u2019s a system language, the fact that it\u2019s highly efficient and so forth is incredible. So Rust, to me, can both be the most repugnant programming language ever devised for human consumption and a wonderful platform for agentic engineering. These two truths can coexist easily in my head.\n \n \n Lex Fridman \n (00:50:24) \n Do you think that there can be a point at which we can just call agentic engineering programming? Can we just change what programming means? \u2018Cause who\u2019s actually going to be programming the old-school way in a few months?\n \n \n DHH \n (00:50:36) \n I don\u2019t think we should reuse the term because for me, programming implies an understanding of certain primitives, loops, conditions, variables. All the constructs of programming languages are what constitutes programming itself, not the creation of programs. Because you could say in the pre-agentic era, well, your CEO is programming. He\u2019s hiring a bunch of programmers, he\u2019s telling them what to do, and out comes software that he can sell.\n \n \n DHH \n (00:51:05) \n Eh, I don\u2019t think most people would call that CEO a programmer, and I don\u2019t think we should call the vibe coder a programmer either. And vibe coding, if we define it here, is you tell an agent to build software for you. You do not look at the implementation. That, to me, is what separates vibe coding from programming or, let\u2019s say, agent-accelerated development.\n \n \n Lex Fridman \n (00:51:26) \n But don\u2019t you think, so to push back a little bit, don\u2019t you think a programmer, old-school programmer, you, doing agentic engineering is a different kind of agentic engineering than a non-programmer doing agentic engineering? What I mean is, if you know what a for loop is, if you know what function programming is, if you know some of the basic good principles of software engineering, the way you do natural language-based agentic engineering will be different and more systematic, and the scale and the variety of things you can actually build is much bigger than a vibe coder.\n \n \n DHH \n (00:52:02) \n I\u2019m just gonna extend the pro case here a little bit just for the sake of the argument.\n \n \n \n \n DHH \n (00:52:07) \n I actually think for a while it was to my deficit to know as much as I know about programming because I was instructing the agents to do things as I prescribed them to do, and they were very good at that. And in the first agentic moment, let\u2019s not call it an era, it was something that lasted three, three months. Moment\u2026 In the first agentic moment, that felt very productive. I could get more productivity out of telling the agents what to do in the way I would\u2019ve done it, and I used all of my experience as a programmer to do just that. Then I was a little late on the next moment, and the next moment allowed me, allowed anyone, to describe outcomes- \u2026 to describe problems to the agents and get better solutions than if you had a programmer prescribe the path.\n \n \n Lex Fridman \n (00:53:05) \n See, okay, so we\u2019re having fun arguing here. Well, so are you suggesting that there\u2019s problems for which programmers are worse than non-programmers at agentic engineering?\n \n \n DHH \n (00:53:16) \n 100%. And the reason I say that, let\u2019s define it here. The reason I say that is there\u2019s a lot of programmers who are not very good product managers. Software is product management. What should it do? Who should it do it for? How should it do it? How should it look? What\u2019s our priorities? What do we start with first? What does version one include? All of those skills are not easily or equally distributed across all programmers. And in the agentic era where you are letting an agent, an AI, do the implementation, these are the skills you need. And I happen to sit in both camps, and this is why I love this moment.\n \n \n DHH \n (00:53:58) \n I still get to live a little bit in the old world where I look at the full implementation because there are domains and areas where I feel that still brings value, and then I also live in the future. And as I said, when I created AmaWrite, I\u2019ve not looked at a single line of that C++. I actually made it a point to myself that I was gonna treat it as an experiment. This is 100% a black box. I will treat it as though I was any other user who had opinions about how their digital typewriter should work, and therefore felt like it was a fair experiment as if a writer of any other sort who had opinions about how software should be done and how it should work.\n \n \n DHH \n (00:54:39) \n As you mentioned, video editors who know exactly where they don\u2019t like what Adobe Premiere\u2019s doing and how it could be better, but don\u2019t have the capacities to improve it- \u2026 how that could be.\n \n \n Lex Fridman \n (00:54:50) \n The pushback, though, I think what programmers do well, good programmers do well, is systematic design, like think in terms of systems- \u2026 and rigor. And I think still, and probably for a long time to come, even natural language-based control of agentic systems will require a certain kind of rigor. Not overly structured rigor, but really explain very specifically the goals of the system, how to do the verification, what kind of security testing. You wanna do all those kinds of things. Although then to push back on that, the system will\u2026 should be smart longer term.\n \n \n DHH \n (00:55:28) \n I would\u2019ve said the same thing six months ago.\n \n \n Lex Fridman \n (00:55:30) \n Yeah, so this has changed.\n \n \n DHH \n (00:55:30) \n And I think the same thing would\u2019ve been true six months ago. What I have found since, especially even just the last few weeks of the agentically accelerated development of Mach-E, is that more often than not, I have the humility to recognize that the agent knows best.\n \n \n Lex Fridman \n (00:55:49) \n You can legitimately say, \u201cMake sure it\u2019s secure.\u201d\n \n \n DHH \n (00:55:52) \n Correct. And it would know more about what that entails. Now, you\u2019re laughing-\n \n \n Lex Fridman \n (00:55:59) \n What a crazy world.\n \n \n DHH \n (00:55:59) \n \u2026 but it\u2019s true. And in fact-\n \n \n \n \n DHH \n (00:56:02) \n \u2026 a good parallel to this is the AGENTS.md/CLAUDE.md files. There was a hot moment where it was all the rage about micro-optimizing that. Oh, you tell your agent this-\n \n \n DHH \n (00:56:14) \n \u2026 you tell your agent that, and you ended up with this huge file of instructions for your agent. One of the things that Boris, working on Claude Code, shared in an interview recently about Opus 5 was that the system prompt that they ship for Opus 5, and presumably also Fable, shrunk by 80% because the agent not only needed far less human instruction, it was actually being damaged by overly prescriptive humans. And any programmer who\u2019s had a pointy-haired boss knows exactly what that is like. When the boss walks into the room, doesn\u2019t know anything, starts telling you how to program, how to code. What do you do? You sulk. You write shittier code if you\u2019re mandated to do things that are against your better judgment. Why would an agent not be the same?\n \n \n Lex Fridman \n (00:57:08) \n Actually, to push back against myself, I do find the actual skill I\u2019m currently developing, which along those lines is probably what everybody\u2019s trying to develop, is not to overspecify things, is to get out of the system\u2019s way.\n \n \n \n \n Lex Fridman \n (00:57:20) \n Is to specify enough to provide a kinda high-level vision and guidance.\n \n \n DHH \n (00:57:25) \n It\u2019s even better because you don\u2019t even need that. The fundamental insight of modern software development came from the agile software development movement. In the late \u201990s, early 2000s, a group of smart, honest, and brave people came together and said, \u201cThe way we\u2019ve been trying to do software development for the last 40 to 50 years has not worked, will not work. We\u2019ve been trying to specify upfront what software should look like. We thought that we could ask humans what they wanted, write it all down, apply our systems thinking and our rigor to a spec sheet, hand that spec sheet over to a group of programmers who would then implement just that, and everyone would be happy.\u201d Well, guess what? No one was happy. Because no one knows what they want until they receive it.\n \n \n DHH \n (00:58:13) \n You don\u2019t know what a program should do until you play with it. So in the agentic age, you should resist the temptation to be overly specific upfront. Be as vague as you can to manifest something, then interact with the something. The way you arrive at good software is you write a little bit of software, and then you try to use it. It is in the process of using software that you discover what you really want. It\u2019s where you discover what\u2019s important and what\u2019s not important. And agents are incredible at allowing you to bumble into this experience of realizing what you want because you don\u2019t know what you want.\n \n \n Lex Fridman \n (00:58:51) \n Mm-hmm. And I\u2019ve actually done something similar to what you suggested, which is implement different designs and totally different implementations, and then I create HTML pages for myself, PHP pages, where I vote on what I like more and so on.\n \n \n \n \n Lex Fridman \n (00:59:06) \n And it does this iterative thing, and it\u2019s very nice and fun. The interaction with the agentic system, if you create a nice interface for yourself, it can be super fun, and you\u2019re focusing on the design part and the most fun parts of the design part.\n \n \n DHH \n (00:59:18) \n Correct. And this is what humans are really good at.\n \n \n \n \n DHH \n (00:59:21) \n Differential evaluation. You give me three options, I pick one of them. Now, humans actually fall off a cliff if you give them 22 options. That\u2019s the paradox of choice. But you give them three options, they very quickly within a split second will tell you what they like. There are all these ways that exploit the fact that humans will make these snap judgments that are actually pre-intellectual, right? Like, they\u2019re coming from the gut, and then they arrive up in the brain, and the brain tries to rationalize why the gut said what it said. And if you\u2019re willing to let go of some of that intellectual pre-processing rationalization and simply let your gut drive, the agentic age is a revelation.\n \n \n The end of manual programming\n \n Lex Fridman \n (01:00:06) \n So this is a fascinating thing to ask you because you have sort of famously for a long time talked about, like, detailed chiseling of beautiful Rails Ruby code, and now you have switched in a matter of months to not doing that. So, like, what is your current setup look like? \u2018Cause you\u2019re, I assume, are still looking for the beauty, for the chiseling, for the crafting, but you\u2019re just crafting in a different medium.\n \n \n DHH \n (01:00:35) \n And at different levels of abstraction. When I\u2019m working in Ruby code, and we have a lot of Ruby code because it\u2019s in our entire business, and I\u2019m asking agents to make changes to that code, I still sweat the details. What I\u2019m coming to realize is that the economic payoff of that sweat is diminishing rapidly. The reason why I, for 25 years, was sweating every line of code so judiciously was because I knew the payoff of keeping an architecture coherent and malleable was software that could change and evolve-\n \n \n DHH \n (01:01:16) \n \u2026 quickly with a small team, and not exorbitant cost, and not introducing a bunch of bugs when you change one thing over the other. That was the driving economic argument for why you should write beautiful code, \u2019cause beautiful code is easier to understand, it is simpler, it is more malleable. That was premised on humans doing the modifications. I think it is an open question to which degree this still matters. Now, it does matter, and the reason I say that, at least for the moment, is that tokens are still scarce. At this moment in time, we are all token limited. Well, not all, but anyone who doesn\u2019t have endless budgets are token limited.\n \n \n DHH \n (01:02:03) \n So therefore, there is great payoff to writing systems that agents have an easier time dealing with and evolving without having to relearn the entire context. Just like humans, if they can make iterations and changes to the code base without wrecking the architecture, they can make the next change just as cheaply as the last one. And this is the classic ball of mud where you end up with a system that\u2019s a ball of mud-\n \n \n DHH \n (01:02:31) \n \u2026 because it\u2019s just put together in a way where nothing is connected and it\u2019s just a real mess, right? I\u2019ve seen that with agents, and I\u2019ve seen it in our own code bases where, again, the first PR is, like, mediocre of quality, and then if you add another PR on top of that, and then five more down the line, it\u2019s not very good, right? So there\u2019s still a payoff to that, but that payoff is premised on our current moment. I am now able to\u2026 And this is, by the way, where the AI psychosis really comes in, when you try to extrapolate what nine months from now it\u2019s gonna look like. What two years from now it\u2019s gonna look like. But as an intellectual experiment, I think the first computer I used was a Commodore 64.\n \n \n \n \n DHH \n (01:03:11) \n It had one megahertz CPU and 64K of memory. Everyone who wrote software for that machine internalized certain constraints. Well, all the constraints. That machine was nothing but constraints. If you put that programmer in a time machine and teleported him or her to 2026 and gave them a modern computer today and said, \u201cNow write me a piece of software,\u201d they\u2019d be lost for a little bit because all their techniques and all their heuristics would simply be wrong. Or not wrong, because efficient software is still beautiful. They would be out of date with the value that they could create. The amount of optimization you have to apply for a one megahertz computer to produce a video game is just very different from what you have to apply today.\n \n \n Lex Fridman \n (01:03:59) \n Just imagine that guy that was programming back then on a Commodore. Just remember, remember all those, like what programming felt like and what it meant, highly constrained resources- \u2026 how slow everything is.\n \n \n DHH \n (01:04:11) \n Beautiful in many ways. Beautiful.\n \n \n Lex Fridman \n (01:04:12) \n It\u2019s chiseling. But-\n \n \n \n \n Lex Fridman \n (01:04:15) \n \u2026 imagine that person coming now.\n \n \n DHH \n (01:04:17) \n It\u2019s romantic too, though. Let\u2019s also imagine that. This is one of the things, whenever we bemoan modern car culture, for example, right? And then we think like, \u201cOh, wasn\u2019t it better when we all had the horse-drawn carriages and so on?\u201d No, it wasn\u2019t. Do you know what New York smelled like when we used horses for transportation? It was an open sewer. They just shat everywhere.\n \n \n Lex Fridman \n (01:04:38) \n Yeah, but have you played Red Dead Redemption? You know how cool those cowboys were-\n \n \n \n \n \n \n DHH \n (01:04:43) \n \u2026 to experience the nostalgic past, when you don\u2019t get all the furry impulses and senses.\n \n \n Lex Fridman \n (01:04:50) \n So basically programming, old school programming by hand is kinda like the cowboy culture that we romanticize and make movies about.\n \n \n DHH \n (01:04:57) \n It\u2019s already happening, the romanticization of handwritten code. And I have some of it because it was a very romantic era. I\u2019m grateful to have been alive for 20 years of economically valuable handwritten code. That was a, that was a good time.\n \n \n Lex Fridman \n (01:05:16) \n Where handwritten beautiful code was also valuable.\n \n \n DHH \n (01:05:20) \n Correct. Because handwritten beautiful code exists as much today as it did yesterday. There are people who willfully write new video games for the Commodore 64, or for the Sega Mega Drive, or for other vintage consoles. In fact, one of my favorite video games of late is ModRetro\u2019s. Do you know Palmer Luckey\u2019s- \u2026 side gesch\u00e4ft, which is recreating old consoles. So he\u2019s recreated the Game Boy.\n \n \n \n \n DHH \n (01:05:51) \n It\u2019s amazing. Look it up. It\u2019s really cool. But what\u2019s even cooler than the fact that he recreated this hardware and the original chunkiness, but brought it up to date with sort of modern screens and so on, now they just put out the Nintendo 64 too. Incredible. Super cool. Wild attention to detail, input lag, and all the other things. But what\u2019s also very cool, that one, I have to wave. That one right there, and hover over it, \u2019cause then you get to see what I\u2019m talking about. That is the Chromatic-\n \n \n \n \n DHH \n (01:06:21) \n \u2026 Tetris re-implementation.\n \n \n \n \n DHH \n (01:06:23) \n I played a lot of Tetris on the original Game Boy. Probably one of my top three favorite games of all time. And they rewrote it with one change. If you hit up, the brick slams. And that speeds up the game by about 400%. It is an incredible game, and it\u2019s a new Game Boy game.\n \n \n DHH \n (01:06:47) \n I think the Game Boy came out in \u201989 or \u201988. It\u2019s a very old machine. The fact that there\u2019s still people writing new old games is awesome, and the reason they do it in part, now here they\u2019re doing it because they want Tetris, but there are people who do it just for the love of the constraints, for the love of the romantic notion of writing that. Just like there are people still riding around on horses, not to get places fast, but because they like the mode of transportation where it\u2019s a literal biological being below you propelling you forward.\n \n \n Lex Fridman \n (01:07:21) \n And also, we should say that that beautiful code is great training data. So in some sense, the beauty that we\u2019ve built over a few decades-\n \n \n \n \n \n \n DHH \n (01:07:31) \n We gave birth to this moment. What a privilege, and I feel that very personally because I\u2026 Almost all of the code I\u2019ve ever written is public code. The vast majority of my career has been dedicated to writing open source code, and therefore, some of that code, some of those beautiful lines are in the training set. And in fact, I\u2019ve heard people do this when they write Ruby code. They ask the agent, \u201cWrite it like DHH would.\u201d\n \n \n DHH \n (01:08:01) \n And they\u2019re pleased with the result, and that does warm my heart a little bit, that I helped give birth to this moment in my tiny little small part. And anyone else who contributed open source code or any code at all that the agents have had access to over the last, well, entire time programming has been around as a discipline, is now part of these new agent overlords.\n \n \n Lex Fridman \n (01:08:22) \n So you don\u2019t have a little bit of sadness that writing beautiful code by hand is now less and less useful?\n \n \n DHH \n (01:08:29) \n Am I sad that if I had to get a job tomorrow, it\u2019s unlikely that I could apply my skills in such a way that I would be paid to write these manual lines of code? No. I did it for 25 years. Like, I wanna see something new here, especially if we\u2019re all gonna live forever, which I hope we don\u2019t. Maybe we\u2019ll get to that. But, like, I\u2019ve done that. That\u2019s enough. That\u2019s fine. I don\u2019t feel any more nostalgic about that than I do about the fact that I don\u2019t have to spend my time in the field with a hoe or at the assembly line.\n \n \n Lex Fridman \n (01:09:01) \n Well, hold on a second. You are one of the best programmers in the world at writing handcrafted beautiful code, appreciators of beauty with great taste.\n \n \n \n \n Lex Fridman \n (01:09:11) \n That has been replaced.\n \n \n \n \n Lex Fridman \n (01:09:14) \n Yeah. That has been replaced. Like, you have now\u2026 You\u2019re just fast evolving, and you\u2019re able to\u2026 One of your really great qualities is you\u2019re able to change your mind and evolve very quickly. But, like, you had to kill that other person.\n \n \n DHH \n (01:09:30) \n I am grateful for every moment that brought me to where I am right now, and therefore, I have also tried, and maybe this did take some practice. I will concede that. Maybe I wasn\u2019t always like this, but stoic philosophy, amor fati, loving your fate, is an incredibly liberating way to live. And you could attack that and say, \u201cWell, that\u2019s a point of privilege.\u201d True, but also I had the same directional point when I didn\u2019t have all this, quote unquote, \u201cprivilege.\u201d So I think there\u2019s a way to live and interact with the world where you accept the things you can change and the things you cannot change and fall in love with all of it. Fall in love with the fact that the world today has jumped two decades forward into technology industry from where it was last year.\n \n \n Advice for programmers\n \n Lex Fridman \n (01:10:24) \n So can we talk about the general anxiety that programmers feel going through the same transformation? They\u2019ve maybe gone to university, majored in computer science, dreamed of being programmers and building stuff, high salary, and now everything\u2019s changing, and there\u2019s deep anxiety about, \u201cWhat do I do when\u2026 with my life?\u201d Can you empathize with that, and what the hell are they supposed to do?\n \n \n DHH \n (01:10:51) \n Hugely. Because just because I have this disposition, I\u2019m keenly aware that that\u2019s not evenly distributed, that not everyone sits with this position and this optimism for the future. But I do think I wanna separate things a little bit. I think you\u2019re gonna have a hard time coping with the new reality if the only thing you loved about programming was the mechanical bits of putting the right logical constructs together to produce something other people told you to produce. Because that mechanical process is under threat. If you are, as you just mentioned, excited about building things, I don\u2019t think you\u2019re under threat at all. In fact, I think there\u2019s a great argument for us needing far more builders than what we have now, and if you look at the employment stats, it\u2019s fuzzy.\n \n \n DHH \n (01:11:53) \n It\u2019s not clear what\u2019s gonna happen at all. Some stats actually show an increase in openings because the advent of AI is so dramatically lowering the price of programs that people want a lot more programs. This is the classic, the Jevons paradox, that says when the price of something goes down, there\u2019s gonna be more demand for it. And this is the example of the ATMs too. When the ATM originally came, a lot of bank tellers were very afraid for their job because suddenly there was a machine that could dispense money from people\u2019s account. Well, what the ATMs did was lower the price of a branch, and suddenly banks could afford to open a lot more branches, and we ended up with more bank tellers than we did before. Now, none of this is guaranteed. There are also moments where things do change.\n \n \n DHH \n (01:12:41) \n The vast majority of people worked in the fields up until the late 1800s, and then suddenly we got mechanized harvesting tools, and we did not need human labor with a hoe and an ox out there. We had mechanical tools to do that for us. In that moment, are there people who were threatened that their job and their livelihood was at stake? Yes. Was that a reasonable thing to feel threatened about? Yes. The Luddites smashing the weaving machines-\n \n \n DHH \n (01:13:13) \n \u2026 in England had the same thing. And I think maybe that\u2019s the better parallel than people working the fields, because these were actually highly skilled professionals doing a job that they liked on rather favorable conditions, not slaving outside in the sun all day. But where we live now, would we still like for clothing to be this heavily constrained resource? Like, what if this T-shirt was $400? I mean, okay, then I\u2019d have two. Well, my wardrobe probably looks like I have two. But I think the general progress of mankind depends on productivity improvements. And what are productivity improvements? This is very important, because in the abstract, I think everyone says, \u201cYay, productivity. That\u2019s good.\u201d What did you think, this was all just vibes? No.\n \n \n DHH \n (01:14:04) \n Productivity means fewer people to do the same number or the same job. Now, the amount of job you want done may increase, and therefore you get more people, but it also may not. There may be pockets where a company just needs a certain set of fixed tasks done, and suddenly they can do them with a 10th the number of people. This, by the way, could\u2026 is tragic and difficult in the moment for the individual being laid off. It\u2019s also amazing for the economy at large. Suddenly you\u2019ve freed up these resources who can now go do more productive things. This is how the whole economy evolves and improves, and how we get growth, and growth is good. I think this is the other thing to make a strong defense for, that growth in general is a good thing.\n \n \n DHH \n (01:14:54) \n We do not want to wind the clock back to 1920 or 1950. We should be excited about the fact that we\u2019ve discovered new technology that allows us to do vastly more. And while some of the tasks that may be taken over by AI were tasks that humans were excited to do, and they\u2019re just no longer economically viable, a lot of the tasks were just drudgery. This is certainly true of my programming. There were moments of programming I thought were amazing. These were the moments that produced that state of flow we talked about last time, and they were ecstatic. How often did they happen out of the course of a year? If you took your average programmer, how much time out of 2,000 hours at the job did that person spend in a flow state?\n \n \n DHH \n (01:15:47) \n 100? 200? 25? I think all those answers could be plausible for someone. And taking that drudgery, handing it over to machines, that is the history of civilization.\n \n \n Lex Fridman \n (01:16:03) \n Yeah, I mean, the hours spent debugging. So programming has so many painful, painful hours in it. And now they\u2019re gone. For the\u2026 I mean, for the most part, it\u2019s just fun.\n \n \n DHH \n (01:16:16) \n A lot of it is gone in a lot of domains, and the rest of it may be gone very soon. So that\u2019s where some of the romanticism also comes in- \u2026 because pain in the moment is difficult. It\u2019s hard. It\u2019s annoying. Pain in the rearview mirror is accomplishment, proud, learning, steps forward, all the things, right? So we encompass all of this as part of the human experience, that we hate pain in the moment, but we cherish it when it\u2019s five minutes past and we don\u2019t remember the hardship. We just remember the progress.\n \n \n Lex Fridman \n (01:16:51) \n So certainly at the societal level, at the macroeconomic level, growth is exciting. But at the individual level, there\u2019s going to be a lot of anxiety, potentially a lot of suffering. So by way of advice, if you\u2019re like a young DHH or young developer programmer now, what would you, what would you advise they do?\n \n \n DHH \n (01:17:11) \n Don\u2019t try to anticipate anything. You will literally go crazy. Because even the smartest brains in the business cannot anticipate what two model hops from here is going to look like. It\u2019s an absolute waste of time, and you will develop an AI psychosis trying to deduce what two years from now is gonna look like. Focus on right now, and right now is the most incredible time to be into computers. You can make them do the most amazing things if you lean in. If you maybe for a hot moment force yourself to learn where the state of the art is, I double dog dare you not to get excited about what\u2019s possible.\n \n \n Lex Fridman \n (01:17:57) \n And would you recommend building publicly?\n \n \n DHH \n (01:17:59) \n I think it\u2019s optional actually to build publicly. I think it\u2019s, it\u2019s nice to do because then you get to be part of a community, and there\u2019s some camaraderie in open source. Not some camaraderie. There\u2019s an enormous amount of camaraderie. Open source is an incredible source of community and camaraderie, and it\u2019s also a great way to stem some of that personal existential dread. If you\u2019re amongst other humans, you can get infected with good forms of mind viruses, exciting forms of mind viruses, the ones that tell you that it\u2019s all gonna be okay, and the future looks bright, and if we build a bunch of things together, we can do things none of us ever dreamed we could do.\n \n \n DHH \n (01:18:43) \n I was just talking to Ryan Hughes, who\u2019s been my main partner on Omarchy almost since day one, and we were literally talking about yesterday how when we started on this journey together, our ambitions were very modest. We\u2019re just like, \u201cCan we get this distro to not crash at 3:00 AM because of an Arch package that just got pushed out outside of our control? That would be amazing. Then it\u2019d be great.\u201d Now we\u2019re leaning back and thinking, \u201cHow can we take over the world? How can we make Omarchy go to Mars? How can we do everything we\u2019ve ever dreamed of?\u201d And that level of ambition inflation is partly because things have gotten better with agents, but hugely also because we\u2019ve surrounded ourselves with others excited about the same journey.\n \n \n DHH \n (01:19:33) \n And if you\u2019re just sitting in your little isolated cave worrying about the future, yeah, you\u2019re gonna go a little nuts. I think if we learned anything during COVID, it\u2019s that people will go nuts sitting in their little cave inside by themselves worrying about what\u2019s gonna happen. That is indistinguishable from depression, ruminating endlessly about things you can\u2019t control. I\u2019m sorry to channel some Jensen Huang here, but that\u2019s just loser talk. You don\u2019t have to be a loser. You can choose to lean in and- \u2026 win. And win is very broadly defined as learning more, making more, contributing more, being part of more. And at the end of the day, what are your choices? You don\u2019t have a choice, mate.\n \n \n DHH \n (01:20:29) \n The future\u2019s coming whether you like it or not, so you might as well choose to be excited about it.\n \n \n Lex Fridman \n (01:20:34) \n But I mean, we should emphasize the fact that planning is nearly impossible with AI. I mean, most of us plan our life a little bit. When we\u2019re younger, you have, like, hopes and dreams. There\u2019s a reason you go to college. There\u2019s a plan underlying that.\n \n \n \n \n \n \n \n \n \n \n \n \n Lex Fridman \n (01:20:51) \n \u2026 nuts that it\u2019s impossible to really plan. Because, like, you really don\u2019t know if there will even be a Claude Code, a Codex, and Cursor six months from now. Maybe it\u2019ll be you\u2019ll be talking to WhatsApp. Maybe it\u2019ll be all open call type. Maybe it\u2019ll just all be voice. Maybe Omarchy will take over the world and it\u2019ll be voice. Maybe there won\u2019t be operating systems anymore. You\u2019ll just\u2026 There\u2019ll be this glowing orb that just you just talk to because if natural language is enough to accomplish all tasks in life, like, what is life about? Like, what are the major tasks you need to accomplish in life? If you can just do that with voice, and it writes all the programs for you, and all is systematized, and what are the jobs needed then? Maybe it\u2019s more about the service industry.\n \n \n Lex Fridman \n (01:21:38) \n Maybe there won\u2019t be programmers, like agentic engineers at all. And so yes, your advice is wise to focus on the moment, but that doesn\u2019t mean that it\u2019s just a crazy rollercoaster ride, \u2019cause we don\u2019t know what this December or what this January will look like.\n \n \n DHH \n (01:21:55) \n We don\u2019t know, and therefore, we should choose to have faith that it\u2019s gonna pan out. That is my best advice towards fighting anxiety, is that you have a choice, to a large degree, of how to lean. Do you wanna lean into PDoom? PDoom is gonna happen whether you lean into it or not. You don\u2019t have control.\n \n \n Lex Fridman \n (01:22:22) \n But there\u2019s also actual daily schedule of how much learning are you doing, how much building are you doing, how much diversification you\u2019re doing. And for example, I traveled across rural China recently, and I did it quite deliberately because you\u2019re talking to people in China, in rural anywhere, where time moves much slower, where there\u2019s no discussions about agentic engineering. And the ti- and I started by going on the Great Wall of China. That\u2019s timeless. That spans centuries. And, you know, there was, like, this kind of tension of FOMO, like maybe I\u2019m missing out on something.\n \n \n Surviving Internet Hate\n \n Lex Fridman \n (01:23:02) \n But then I very deliberately wanted to find the timelessness of human existence, really plug into the fact that, okay, it always feels like everything\u2019s changing throughout human history, but, you know, the universals are still true of what matters in life. And I just wanted to make sure to keep connected to that.\n \n \n DHH \n (01:23:22) \n You\u2019re not gonna miss anything. This is the part that actually grinded my gears in the early phases of this transition. Everyone was so up in arms that if you didn\u2019t\u2026 It\u2019s loops now. Oh, no, no, we\u2019re done with loops. It\u2019s graphs now. Oh, no, no, we\u2019re done with that. It\u2019s harnesses this, right? They\u2019re constantly churning through the frontier, which in one way is actually very exciting. This is what happens when a new field opens up. It\u2019s like a new portal, and it\u2019s all spilling out. That\u2019s exciting. But it also means that if I had just been backpacking for the last year, hadn\u2019t touched a computer, hadn\u2019t witnessed this agentic moment, and I just showed up yesterday, do you know what? I would\u2019ve been caught up in two weeks.\n \n \n DHH \n (01:24:08) \n There\u2019s not any accumulation, which is in some ways a great relief. If you missed the past year, you can catch up to the frontier in two weeks. If you\u2019re a programmer who was out hiking the Himalayas for a year and you come back, you can catch up in two weeks. And this is actually the great credit to the progress we\u2019re experiencing. We have so many experiments running simultaneously right now that are constantly and ruthlessly sorting what works, what doesn\u2019t work. You don\u2019t have to remember or even be part of that entire journey. You can just show up for the results.\n \n \n Lex Fridman \n (01:24:45) \n But the key step there, when you come back from the backpacking journey, is to be willing to become a totally new, different human because things are changing so fast.\n \n \n \n \n Lex Fridman \n (01:24:56) \n I mean, literally, if you left for the backpacking journey- \u2026 in October last year and came back in April or May-\n \n \n DHH \n (01:25:04) \n You might as well have been in the cryo chamber for 100 years.\n \n \n Lex Fridman \n (01:25:06) \n \u2026 it\u2019s like, \u201cWhat, what do you mean? We\u2019re not programming anymore?\u201d This, it\u2019s a very\u2026 it\u2019s a huge transformation.\n \n \n DHH \n (01:25:12) \n It\u2019s startling, and I actually\u2026 I mean, I\u2019m putting some of this on slightly for effect. I also recognize that it\u2019s okay to grief for a hot moment. That\u2019s not my disposition. It\u2019s not how I do it, but I accept that that\u2019s a time-tested mechanism for coping with a sense of loss. And it\u2019s okay to have a sense of loss that the world that you knew and maybe you were very fond of is no longer the same.\n \n \n Lex Fridman \n (01:25:43) \n I maybe should say out loud and explicitly that I do have that, the Native American with a tear rolling down my\u2026 Like, there\u2019s a sadness. It\u2019s a goodbye. It\u2019s a goodbye to the old world of programming.\n \n \n DHH \n (01:25:55) \n That\u2019s\u2026 it\u2019s so fascinating to me.\n \n \n Lex Fridman \n (01:25:56) \n I\u2019ve spent so many years-\n \n \n DHH \n (01:25:58) \n But it all amounted to this moment. That is, to me, what makes the entire journey so meaningful. AI did not arrive from the sky. It was not alien technology that had blasted through the universe and suddenly showed up on our shores. AI has been around for as long as computer science has been around. In the \u201950s, they thought the problem was gonna be cracked in about a decade, the smartest people at the time. It took a little longer, but the reason it took a little longer was in part because we took some blind turns on the road to AI and neural networks and thought that there were other symbolic representations that were gonna carry the day, and they didn\u2019t. And that probably set us back about 15 years. So some of it was also we had to go through the gaming revolution, 3D games.\n \n \n DHH \n (01:26:47) \n If we had not had Quake, if we had not had Duke Nukem, if we had not had Unreal Tournament, we\u2019d never have gotten AI because we would never have gotten the GPUs, and therefore we would never have been able to get AI in the shape it is now. So everything we did up until this moment was required to get there. You were part of that. Holy\u2026 What a blessing. What a blessing to have contributed in your small part, either by training or by playing video games. Literally, you could just choose to think of it that way. All those times you spent in, for my case, in the \u201990s just playing video games, they were-\n \n \n Lex Fridman \n (01:27:23) \n You were contributing.\n \n \n DHH \n (01:27:24) \n Yeah, I was contributing to the AI revolution right there and then moment.\n \n \n Lex Fridman \n (01:27:27) \n To the progress of human civilization. No, for sure. Civilization progresses with the death of the old and the birth of the new, and it continues in this way, but it\u2019s still the grieving process is a part of it, I think.\n \n \n DHH \n (01:27:39) \n Take a moment. It\u2019s okay. Take a\u2026 And also, you don\u2019t even have to accept that it\u2019s all dead. Like, if you are still very attached to the hand chiseling, which if you\u2019d asked me a year ago, I would have predicted that I would\u2019ve been more attached. I\u2019m surprised-\n \n \n Lex Fridman \n (01:27:58) \n I\u2019m kinda surprised\u2026\n \n \n DHH \n (01:27:59) \n \u2026 a little that I\u2019ve not been more attached. But the reason I\u2019ve not been more attached, I think, is that I found something more fun. And if it hadn\u2019t been like that, if AI had just replaced the thing I loved and gave me something I hated, I\u2019d probably be a little bitter.\n \n \n Lex Fridman \n (01:28:16) \n I mean, this really feels\u2026 It\u2019s\u2013 it\u2019s hilarious and awesome to watch and inspiring to watch. It\u2019s like Picasso all of a sudden f\u2013 like, getting access to image generation and getting super excited and switching in a matter of months. That\u2019s literally what happened. So, like, it\u2019s an artist that appreciated beautiful code, and all of a sudden you\u2019re talking about having fun-\n \n \n DHH \n (01:28:37) \n I think Picasso is actually a good parallel because if you look at his early work where he learned his craft, it was painting realistic paintings, right? Like, he was training to be a master of the old ways of depicting reality the best way we knew how, and then along comes all these other forms of expression. Cubism comes along.\n \n \n DHH \n (01:29:01) \n And all these o\u2013 forms of abstract expression, he\u2019s not decrying that he\u2019s not doing Renaissance paintings anymore, that he\u2019s not just painting the perfect depiction of an apple. He\u2019s reimagining the apple to be a freaking square, and he\u2019s getting excited about that. And I feel that transition, that we\u2019ve gone from this one mode of expression, and we\u2019ve suddenly opened the gates where the, like, the color spectrum is so much wider. I\u2019m seeing more. And it\u2019s perhaps because the way I became a c\u2013 programmer. I did not become a programmer because of a deep love of if statements. I became a programmer because I wanted programs. I wanted things to exist that did not exist. And at first, I had to learn how to program to make that happen.\n \n \n DHH \n (01:29:56) \n Then I happened to fall in love with programming as a craft, and then I spent two decades really diving deep on that. But now I\u2019m back. I\u2019m back to having an idea and being impatient beyond belief to see it exist in the world, and AI has just shrunk that down to almost nothing. So it\u2019s an acceleration of a rediscovery. Now, I understand that that\u2019s not true for everyone. There are plenty of programmers who just loved programming right from the beginning, and they were attracted to the logical constructs and so forth. I mean, I\u2019ve been there for two decades, so I get it. But I also think that you\u2019re able to discover new sides of yourself-\n \n \n DHH \n (01:30:41) \n \u2026 if you let it in. Again, I keep coming back to this almost regret minimization framework, to borrow Jeff Bezos\u2019s terms here. Are you gonna look at yourself two years from now and think, \u201cOh, I spent my time well being pissy about the present, about the fact that my industry changed\u201d? Or would I two years from now look back upon this moment and be proud of myself for leaning in, for maybe having a moment of grief? Again, there\u2019s room for all of it. And then going, \u201cI\u2019m gonna learn how the world spins now, and I\u2019m gonna make not just the best of it, I\u2019m gonna make more of it.\u201d\n \n \n Lex Fridman \n (01:31:22) \n I mean, it is true, a lot of programmers, a lot of friends of mine, me, obviously you, are having fun.\n \n \n DHH \n (01:31:30) \n I have had more fun with computers in the last three months than at any time previously.\n \n \n Lex Fridman \n (01:31:38) \n Quick bathroom break, and then I gotta ask you about how you actually interact with the systems now.\n \n \n \n \n Programming setup for AI Agents\n \n Lex Fridman \n (01:31:46) \n I gotta ask you about how has your setup programming setup changed? So, keyboard, voice, what\u2019s the IDE?\n \n \n DHH \n (01:31:57) \n It\u2019s crazy to think about now, but yeah, I used TextMate for almost 20 years. I used TextMate starting in 2005, I think. I helped get the first version out, and then I just wasn\u2019t interested. I wasn\u2019t in the market for an alternative. And it wasn\u2019t until the switch to Linux that I was forced out of my habitat. And now with the switch to, what are we calling it? Agentic engineering? Oh, I fucking hate that term. We gotta come up with something that sounds as plain as programming, but encapsulates the fact that it\u2019s with agents. But-\n \n \n Lex Fridman \n (01:32:35) \n I still think it should be called programming at this point.\n \n \n DHH \n (01:32:36) \n \u2026 All right, let\u2019s just call it programming. Programming with agents requires a different tool set. It really does. And the main change here is that you\u2019re going from single thread programming in your head to parallel processing. When I was writing code, chiseling it by hand in TextMate or even Neovim not that long ago, I would just focus on one problem at the time, and I would methodically work my way through it, and that was actually the portal to flow. The portal to flow was deep immersion into a single problem, see it through to the end.\n \n \n DHH \n (01:33:15) \n That\u2019s not how it works with agents, in part because the agents are at once both too fast and too slow. They don\u2019t give you an immediate reply on something that you asked them to do that\u2019s the same as typing on a keyboard. So you have to let the agent cook- \u2026 for a bit, and therefore you realize, well, if I just sit around waiting for them, first of all, that doesn\u2019t feel productive. Even if the agent, just one of them, can be highly productive, it does not feel productive. It does not feel good. It feels actually like you\u2019re a little bit useless.\n \n \n DHH \n (01:33:50) \n And maybe I had a moment when the first agentic moment was there and we, I was running mostly one agent at a time where I felt like, \u201cI don\u2019t know about this.\u201d But you can solve a lot of hard problems by simply throwing more resources at it. This is the whole scaling law of AI itself, right? That if you parallelize these things and you\u2019re not running one agent, but you\u2019re running a handful, you can feel like you\u2019re in a flow state because you\u2019re constantly doing programming work in the sense that you\u2019re making decisions, and you\u2019re helping either unblock an agent because it has a question about which direction to take, or you\u2019re ready for a new task. And to do that, you need a different setup.\n \n \n DHH \n (01:34:35) \n I started first doing it in tmux and just having separate panes and having separate splits. Basically a terminal with tabs is a good way to think about it. You open a bunch of tabs. I think most humans know exactly how that works. They don\u2019t work in just one tab.\n \n \n Lex Fridman \n (01:34:50) \n So still sticking to the terminal, CLI.\n \n \n \n \n Lex Fridman \n (01:34:53) \n So you\u2019re not using Claude Code app or the-\n \n \n \n \n Lex Fridman \n (01:34:56) \n \u2026 Codex app or the-\n \n \n DHH \n (01:34:57) \n I love the fact that this agent revolution was kicked off in the terminal because I was already a huge fan of TUI\u2019s, terminal user interfaces, and the terminal in general. That feels like a really nice place to be. It\u2019s a beautiful place to be. The modern terminal is just a good-looking place to work. So I like that. And then this fact of having multiple agents, especially once it\u2019s not just multiple agents running on your own machine, but you start running multiple machines. Now tmux alone is not enough to keep track of it, and that\u2019s why as of late I\u2019ve switched to this thing called Herdr.\n \n \n DHH \n (01:35:40) \n And Herdr is essentially tmux plus agent notifications. So whenever your agent is done and needs something for you, it goes ding, a little bell telling you it\u2019s ready for its human\u2026 Is it master or servant? I\u2019m not quite sure always. But it is ready for a decision, and it also keeps track of whether it is working mode or not. So I have this Herdr set up. I have multiple Herdr setups actually running on individual machines. I went on this crazy phase just about a month ago realizing that doing this work on a single machine is not fast enough. It\u2019s like I\u2019ve discovered multi-core programming, but I only have two cores. I\u2019m like, \u201cWhat if I had 16 cores? What if I had 32 cores? What if I had 64 cores?\u201d So I instantly went out and I bought these amazing KVMs called GL.iNet Comets.\n \n \n DHH \n (01:36:35) \n And what they do is it\u2019s this little box. You plug in HDMI, you plug in USB and connect it to the computer. It\u2019s like a KVM. So a KVM is a remote way of controlling a computer, but what\u2019s special about this is just how easy it was. You connect this thing in, you go to a webpage, log in once, set one password. Now this thing can hop on your tailnet. This has been the other revolution of the last year for me, is discovering these WireGuard networks. Tailscale is essentially turning all the computers you have into a local network wherever you are. Like right now on my phone, I have direct access to all the computers in my Malibu office. I also have access to all my computers in my Copenhagen office.\n \n \n DHH \n (01:37:26) \n And I can treat them as though I sat right next to them without having to punch holes in a firewall or set up complicated VPNs. And what this does is it just decreases the friction it takes to get new compute online. So as soon as I discovered this, I looked at my closet, and I realized I had a bunch of mini PCs from prior experiments, and I just said, \u201cWhat if I just connected all of them?\u201d\n \n \n \n \n DHH \n (01:37:51) \n And I just connected four of the computers-\n \n \n \n \n DHH \n (01:37:53) \n \u2026 in a closet. They all had their little Comet, and suddenly I could run agents on more computers at the same time, and I could control them all with Herdr. And it did get to a point where I maxed out my own processing power.\n \n \n \n \n DHH \n (01:38:09) \n That I think at about\u2026 What do we want to put it? About four to five machines running, I don\u2019t know, three agents. I have about 16 threads. That\u2019s what I can run. And the faster the agents run, of course, the fewer threads I can run, but at the current pace, I can run about 16 threads- \u2026 at full acceleration. And that\u2019s part of why I\u2019ve gone from just being excited about the agent moment to being delirious. Because I think as you say, some of it is a little assaulting. Again, we\u2019re on\u2014we\u2019re on a dial-up bandwidth-wise with our computer. I mean, it\u2019s actually hilarious. So you think of I\u2019m sitting here a year ago, right? I\u2019m chiseling my code. I\u2019m writing my lines.\n \n \n DHH \n (01:38:55) \n And over the course of, like, an hour, I will have written one beautiful controller, one beautiful model. Like, that\u2019s one file. And I\u2019ve really worked on it, right? So maybe there\u2019s 60 lines left. So maybe my bandwidth at this moment is, like, 30 lines an hour. I think that\u2019s even high. Maybe it\u2019s 20 lines an hour. Now I\u2019m running 16 threads. I\u2019m producing at some times hundreds of lines an hour or hundreds of lines of code per hour. Now, let me pause myself there for a hot moment. I hate that metric, right?\n \n \n DHH \n (01:39:28) \n Like, lines of code is a stupid metric in general to measure these things, and I think it\u2019s right from some in the programming community to ridicule the AI psychosis when everyone talks about the number of lines of code they\u2019re writing, and then you ask them, \u201cWhat did you build?\u201d And then they\u2019re like, \u201cWell, I\u2026\u201d And then no good answer comes out, right?\n \n \n Lex Fridman \n (01:39:47) \n It\u2019s just a nice shorthand to-\n \n \n DHH \n (01:39:49) \n It is a nice shorthand. It is a representation or encapsulation of how much output that\u2019s coming out. Now, whether the output is good or bad-\n \n \n DHH \n (01:39:57) \n \u2026 is not a referendum on that. But I was producing and producing so much more now, right? And therefore I\u2019m able to keep all these threads going. So that\u2019s the setup. It\u2019s still Neovim, but at this point, I\u2019m not writing a lot of code, so I\u2019m using Neovim as a project browser and then as a way to kick off lazygit to see the change log for what\u2019s there. And even that, I\u2019d say if GitHub was a little faster at showing you your pull request, the web is probably actually a nicer place to do that. I don\u2019t know. There\u2019s also this other tool I\u2019ve been playing with a bit called Hunk, which just produces diffs in a really nice way. So you could look at that too.\n \n \n DHH \n (01:40:37) \n But I find that when I look at Hunk, I only see the change set, and when I\u2019m reviewing output from an agent, I often want to see the surrounding context. Oh, yeah, so it changed this file, but actually, what do we have in this other file that wasn\u2019t touched but maybe should\u2019ve been touched? So that\u2019s why I still like Neovim as a way of doing it. But it\u2019s all happening, by the way, of course, in Omarchy. So it\u2019s all happening on Linux. And this was the other major breakthrough with agents. Agents love the Unix philosophy. It loves individual tools that it can invoke through the command line, and there is no operating system on Earth of the majors, I\u2019m counting three here, Mac, Windows, Linux, that works as well with that mechanism as Linux.\n \n \n DHH \n (01:41:25) \n Everything in Linux is either a config file or a CLI tool. Now, that was its main drawback five minutes ago. This was the reason people didn\u2019t like Linux. It\u2019s like it\u2019s all config files and CLI tools.\n \n \n DHH \n (01:41:39) \n What great irony that the universe has played it upon us that now the drawbacks of Linux five minutes ago are now its major selling points. This was one of the things\u2026 I thought I was stuck for a weekend with a Mac four months ago. I thought I had a computer at the place I was going that was a Linux machine, so I didn\u2019t bring my laptop, and I found out when I arrived I only had a Mac Mini. So I was gonna make the best of a bad situation here and set up my Mac in some of the ways I\u2019ve been thinking about with Linux. And you can actually do a lot now. Homebrew has gotten really quite good. Homebrew is the missing package manager for the Mac, and no one has done more to move the Mac forward in terms of ease of setup.\n \n \n DHH \n (01:42:31) \n But still something as simple as Raycast, I don\u2019t know if you\u2019ve used that. That\u2019s the- There\u2019s no config file that you can just access. You have to go into the GUI, export a file, then take that file, I don\u2019t know, in your freaking backpack on a USB key. And then you can import it somewhere else. You cannot automate the entire setup of your machine. You can\u2019t automate at all the configuration of Mac\u2019s default key bindings. That has to be a manual process where you\u2019re clicking with a mouse like a caveman to set up your machine.\n \n \n Lex Fridman \n (01:43:05) \n I mean, there\u2019s ways around it.\n \n \n DHH \n (01:43:07) \n Not good ones. I looked. I tried hard.\n \n \n Lex Fridman \n (01:43:10) \n So here\u2019s my\u2014been my journey. Obviously I\u2019m a Linux person, but because of Adobe Premiere, Adobe products, I\u2019m also a Windows person, so oftentimes I would use WSL, Windows Subsystem for Linux.\n \n \n \n \n Lex Fridman \n (01:43:23) \n So Linux inside Windows, which is\u2026 in the agentic era is not a good kind of Linux, \u2019cause it\u2019s just-\n \n \n \n \n Lex Fridman \n (01:43:29) \n \u2026 it\u2019s a sandbox, and you want the Linux to be unleashed to be able to do everything.\n \n \n \n \n Lex Fridman \n (01:43:34) \n And so I have now woken up to things like GrayCat, and have w- you know, I\u2019m a keyboard person. Where\u2019s the config file? What, can I basically do everything where an agent can, can set everything up for me and save the configuration so I can replicate it across systems and everything\u2019s automated? And so I had to ask a lot of those questions, and a lot of them are missing.\n \n \n DHH \n (01:43:56) \n They don\u2019t have good answers.\n \n \n Lex Fridman \n (01:43:57) \n But oftentimes you could actually just build the app yourself.\n \n \n \n \n Lex Fridman \n (01:44:02) \n Yeah, they\u2019re hacks.\n \n \n \n \n \n \n DHH \n (01:44:03) \n You don\u2019t have to live like this, Lex. There\u2019s a better\u2026 Actually, this is a great moment.\n \n \n Lex Fridman \n (01:44:08) \n Until you have a video editor.\n \n \n Obsessing about speed\n \n \n \n \n \n DHH \n (01:44:13) \n I heard it was your birthday.\n \n \n \n \n DHH \n (01:44:16) \n So I brought a little gift.\n \n \n \n \n DHH \n (01:44:19) \n We are going to get you onto the agentic operating system whether you like it or not.\n \n \n \n \n DHH \n (01:44:24) \n And I asked- \u2026 my friends at Dell whether maybe they had a machine that I could get you for your birthday.\n \n \n \n \n DHH \n (01:44:33) \n So here\u2019s the machine I use- \u2026 which is- \u2026 well, the same one I use, but it\u2019s now yours. It\u2019s a-\n \n \n \n \n DHH \n (01:44:40) \n \u2026 Dell XPS 14- \u2026 already set up with Omarchy 4-\n \n \n \n \n DHH \n (01:44:45) \n \u2026 ready to be configured for you. And once you realize that all your\u2026 By the way, we should almost time this. You can be up and running. You have to answer, like, five questions in about one minute. You could literally do it live if you wanted to.\n \n \n Lex Fridman \n (01:45:00) \n Oh, Omarchy, beautiful, modern opinion Linux by DHH. Yeah, let\u2019s do it.\n \n \n DHH \n (01:45:04) \n Yeah, fill it out while I talk. So-\n \n \n \n \n DHH \n (01:45:07) \n \u2026 exactly what you realized, that pain on the Mac was the exposure I got for about a weekend, and I tried really hard, and I found some decent workarounds, the main one being that Homebrew now has tabs for GUI apps. Because it\u2019s hilarious to think upon the recent past where people would go to websites, like, \u201cI need Zoom.\u201d They would enter zoom.com, find the download link, download a DMG- \u2026 double-click on that, see a weird manual interaction telling them to drag an icon from one place of the screen to the other, and that was how you installed apps. And you\u2019d just do that over and over again until you had all the apps that you wanted, and if you were setting up a new machine and you had, like, 12 apps, you\u2019d go to 12 different websites to download them like a total caveman.\n \n \n DHH \n (01:46:01) \n Homebrew has solved a lot of that, but it has not solved all of it, and has solved nothing in terms of the-\n \n \n Lex Fridman \n (01:46:08) \n Sorry to interrupt, but look how fast this is going.\n \n \n DHH \n (01:46:11) \n Oh, it\u2019s gonna be done in about 40 seconds. And then you\u2019re gonna be inside of Omarchy.\n \n \n Lex Fridman \n (01:46:17) \n Super is the Windows or Command key on your keyboard. So which one is the Super key?\n \n \n DHH \n (01:46:21) \n Yep, that\u2019s the Windows key. Oh, actually, no, no, you have a special edition.\n \n \n \n \n DHH \n (01:46:25) \n There are only about six of these computers in the world.\n \n \n Lex Fridman \n (01:46:27) \n Oh, there\u2019s a Super key.\n \n \n DHH \n (01:46:29) \n It says Super, and look at where the Copilot key is on the other end of the keyboard. That\u2019s the Omarchy logo.\n \n \n \n \n DHH \n (01:46:36) \n This was Spencer Bull at Dell, who I\u2019ve been working with on a lot of this stuff. They\u2019ve been incredible.\n \n \n Lex Fridman \n (01:46:42) \n They, for a moment, let go of the XPS brand.\n \n \n \n \n Lex Fridman \n (01:46:48) \n And I was like, \u201cThis is dumb. This is\u2026 XPS is awesome.\u201d\n \n \n DHH \n (01:46:51) \n It was not just dumb that it got rid of the XPS branding. They also got rid of the Escape key and put in capacitive touch buttons at the top. The mistake that Apple realized like three years ago was terrible, and everyone hated them for them. But now they have an incredible team, and this was actually my first Dell laptop for myself.\n \n \n DHH \n (01:47:12) \n Because I\u2019ve been using Dell on our servers. Dell was what powered our cloud exit. We\u2019ve been running Dell since the mid-2000s, I think. But I never ran their laptops, because compared to a MacBook, they just weren\u2019t up to the task for me. The XPS\u2019 from 2016, or from 2026, with those Intel Panther Lake chips, are the first Dells that I would personally actually use. And it\u2019s funny because when I got mad at Apple originally, I thought, like, \u201cDo you know what? I should just get a Dell,\u201d because we already used them and whatever, and I got a Dell, and I was like, \u201cMan, I can\u2019t do it. I can\u2019t do it. It\u2019s not good enough.\u201d And then lo and behold, this year they just fixed everything. They fixed the battery life. They fixed the performance.\n \n \n DHH \n (01:47:59) \n They fixed the compatibility. They fixed the weight. They fixed the screen. That\u2019s a tandem OLED screen-\n \n \n Lex Fridman \n (01:48:05) \n Yeah, that\u2019s incredible.\n \n \n DHH \n (01:48:05) \n \u2026 which means it\u2019s actually better tech than what you\u2019ll get in a freaking MacBook. It\u2019s lighter than a MacBook, and the chip is finally competitive because Intel spent a good five years developing the 18A process node. It took a long time, but now it\u2019s here, and it\u2019s freaking incredible. We finally have competition to Apple M chips- \u2026 that can go as long as on battery and can have competitive performance.\n \n \n Lex Fridman \n (01:48:34) \n And there\u2019s a beautiful default theme to this thing, and then what is it? Super Space opens up.\n \n \n DHH \n (01:48:39) \n Actually, you can just push the Omarchy key. It opens up the menu. So now you control everything from there.\n \n \n Lex Fridman \n (01:48:44) \n Oh, same thing as the Super. Okay. Cool.\n \n \n DHH \n (01:48:45) \n Same thing as Super Space. So this has been a huge change for me, that we finally have Linux machines that you don\u2019t have to make excuses for. Because I don\u2019t mind making excuses because excuses are really just another word for trade-offs. I love that Framework 13 I think I talked about last time. I still like them. They still make good computers. But I think for a lot of people, they didn\u2019t want a DIY experience. They wanted something that just out of the box felt like it was on par in build quality and whatever with Macs, and now we finally have it. So happy birthday.\n \n \n Lex Fridman \n (01:49:27) \n Thank you so much. An Omarchy Dell laptop from DHH is the best birthday present. This is awesome.\n \n \n DHH \n (01:49:34) \n Check the screensaver, by the way.\n \n \n Lex Fridman \n (01:49:36) \n Yeah, it\u2019s kind of amazing. I\u2019m seeing it.\n \n \n DHH \n (01:49:38) \n On that tandem OLED screen-\n \n \n \n \n DHH \n (01:49:41) \n \u2026 it is absolutely mind-bogglingly awesome.\n \n \n Lex Fridman \n (01:49:46) \n Look at that. There\u2019s like a retro look to the whole thing. It\u2019s like Fiverr pop-\n \n \n DHH \n (01:49:51) \n It\u2019s like running a bulletin board system in- \u2026 the \u201990s, which is exactly what I did when I was 14 years old. So for me, it\u2019s a complete throwback and an awesome aesthetic, combining this sense of modern rising culture in Unix with- \u2026 this throwback of the \u201990s.\n \n \n Lex Fridman \n (01:50:10) \n Yeah, like a retro look for an agent first OS. It\u2019s crazy.\n \n \n DHH \n (01:50:13) \n Which again, pairs\u2026 This was why I was so excited that- \u2026 all these agent harnesses arrived as TUI\u2019s, arrived as CLIs- \u2026 has arrived as a terminal because suddenly, a whole new generation of both programmers and regular users discovered the glory of the terminal. The terminal was actually one of the most impressive and productive user interfaces we can have. Now, we\u2019ve built user interfaces since then that were easier to learn. The mouse is a wonderful invention for letting someone who doesn\u2019t know a lot about computers just click around and find things out, but the terminal is for people who want to learn their systems, who wanna learn the keyboard commands.\n \n \n Lex Fridman \n (01:50:57) \n So you think the TUI is gonna stick around?\n \n \n DHH \n (01:51:00) \n 100%. Now, it\u2019s not gonna be the only thing. Of course it\u2019s not, because these labs now have trillion dollar valuations, so they need to be able to reach billions of people. And I don\u2019t think we\u2019re gonna teach billions of people the intricacies of the TUI. So we\u2019re gonna have other tools, and we already have other tools, which is, by the way, one of the sweet things about Linux, the ChatGPT Codex tool.\n \n \n \n \n DHH \n (01:51:23) \n Literally within two hours of that announcement dropping, I had it wrapped up and ready to go in Omachi. Already now, if you go to\u2026 You can actually hit the Omachi key- \u2026 and then just write ChatGPT. It should pop up as, like, install.\n \n \n \n \n \n \n Lex Fridman \n (01:51:41) \n Sudo password for Lex.\n \n \n DHH \n (01:51:42) \n Yeah, whatever you put in. And then it\u2019s just gonna install it, and it\u2019s gonna pop it up, and you have ChatGPT.\n \n \n \n \n DHH \n (01:51:48) \n And that\u2019s how a lot of the apps are, and all the agentic harnesses are built in that way. Tailscale, as I talked about, if you wanna hop on that, is that way. Dropbox. And the reason all these things are in there is because I use them. It\u2019s because I built this system for me, and I thought, first and foremost, how do I build my dream computer just for me? And if I get my dream computer, I am assured there are other people who are gonna have the same dreams as I do.\n \n \n Lex Fridman \n (01:52:13) \n Now, you\u2019re obsessed about getting the installation to be under one minute, and a bunch of people asked you online why. So why? Why were you so obsessed to getting the installation to be less than 60 seconds?\n \n \n DHH \n (01:52:27) \n First, let me quote Michel Hashimoto, who created Ghostty terminal that\u2019s awesome and also available on Omachi and was also the founder of HashiCorp, who had this great post on X not too long ago that had the killer line, \u201cThe pursuit of excellence deserves no explanation.\u201d Wanting to make something as good and as fast and as beautiful as possible has no need for justification.\n \n \n DHH \n (01:52:58) \n We should want beautiful, fast systems that are a delight to use. Now, the funny thing with the install time is it didn\u2019t actually start with a one-minute mission. I was being modest when it got going. I thought if I could set up my entire system in 15 minutes, it would already be such a dramatic improvement over what came before it that I would be perfectly happy. And it came off the fact that the benchmarking had been skewed so out of whack. You talked about the Commodore 64 being your first computer, right?\n \n \n \n \n DHH \n (01:53:28) \n When you unwrap that Commodore 64 and you hit the on button, in I think about less than one second, the basic interpreter was ready to accept your command. It was literally almost instant. There was virtually no boot time. Do you know how long it took me to set up a new Mac that I had bought for actually the one app that I still haven\u2019t vibe coded my way into on Linux, which is Adobe Lightroom?\n \n \n DHH \n (01:53:53) \n Brand-new computer, out of the box. I set it up. There\u2019s a software update. Before I was ready to install Adobe Lightroom that I had it fully up to date, it took 42 minutes. That\u2019s grotesque. That\u2019s absurd. That\u2019s an insult to everyone who likes computers, that you get a new computer and before you can use it, you have to spend 42 minutes doing updates\u2026 on something that was already installed on it out of the box from the factory. That is just preposterous, and it\u2019s crazy that that\u2019s not even the worst thing. I also got a PC because all these Linux machines, they come as PCs, right?\n \n \n \n \n DHH \n (01:54:35) \n And they have Windows pre-installed. So I got one. This was just three weeks ago. Brand-new machine, Intel Panther Lake, the whole shebang. An hour and 35 minutes from unwrapping the thing until it\u2019s ready to use. How long did you take before you think you were in Omarchy?\n \n \n Lex Fridman \n (01:54:53) \n I mean, it\u2019s less than a minute for sure.\n \n \n \n \n \n \n DHH \n (01:54:57) \n We can build computers like this. We have the technology. We\u2019ve had the technology since like 1981 when the Commodore 64 first came out. The fact that we now have a computer that\u2019s a trillion times faster and you have to spend 42 minutes or an hour and a half setting it up before it\u2019s ready to use is so infuriating to me that it became a passion. And it\u2019s funny, just while we were having a little break here, I was talking to Ryan Hughes, my co-conspirator here. We now have, we think, a line where we can create these special turbo images that are pre-made for specific computers because it relies on some knowledge of the hardware, like the Dell XPS. We\u2019re gonna make a Dell XPS turbo image that can install in about 12 seconds. A full Linux system-\n \n \n \n \n DHH \n (01:55:43) \n \u2026 installed in about 12 seconds.\n \n \n Lex Fridman \n (01:55:44) \n That\u2019s incredible.\n \n \n DHH \n (01:55:45) \n That obsession is what gives me this great satisfaction with using computers, that we do not accept things the way they are. And this is why this agent-accelerated reality that I now inhabit is so invigorating. Because I can look at these kinds of constraints and go, do you know what? I could have stopped when we were at five minutes. That would already have been fast enough. This was the argument on X, right? Why do you keep going? Five minutes is fast enough. Okay, maybe for you, but for me, 12 seconds is a lot more fun. And I always think, try to think these first principles. Okay, in that machine, actually hit the Omarchy key again, and then do speed. You\u2019ll see a list come up, and one of them is, like, disk speed.\n \n \n Lex Fridman \n (01:56:35) \n Speed test, network speed, disk speed.\n \n \n DHH \n (01:56:37) \n Disk speed. Do that one, and then run it. So this is testing the performance of the built-in NVMe drive, and what are you getting?\n \n \n Lex Fridman \n (01:56:45) \n Seven GB a second.\n \n \n DHH \n (01:56:47) \n So seven gigabytes a second. That\u2019s how fast that drive is. Why can we not install a Linux distribution in one second? The Omarchy distribution is 5.8 gigabytes. If we can run at seven gig\u2026 Now, of course, you can\u2019t actually do that because the USB key is a lot slower than that, but you should be able to maximize the underlying physics and protocols that we\u2019re dealing with, and that\u2019s your goal. We gotta get all the way back down to that.\n \n \n Lex Fridman \n (01:57:15) \n Yeah. I love that you\u2019re doing a Linux distribution. \u2018Cause I mean, you could be do\u2026 You could be building anything. But the fact that you\u2019re building a Linux distribution means you\u2019re rethinking about what a computer is, \u2019cause a operating system is the fundamental, like, the foundation of how-\n \n \n \n \n Lex Fridman \n (01:57:32) \n \u2026 you interact with a computer.\n \n \n DHH \n (01:57:33) \n Yes. And it\u2019s been so stale.\n \n \n \n \n DHH \n (01:57:35) \n The macOS we have today is scarcely different from the macOS we had 10 years ago. In fact, it\u2019s worse in many ways because Apple continues to chokehold the control on users. They don\u2019t want them to install software that they haven\u2019t already said good- \u2026 to go for. They don\u2019t want users to reconfigure their hotkeys without doing it manually. They didn\u2019t want to change the 500-millisecond animation switching workspaces. It is an infuriatingly locked-down computer. Now, to Apple\u2019s credit, it\u2019s a pretty good computer for being locked down, but I don\u2019t want a locked-down computer. I wanna own my computer. Better yet, I wanna mutate my computer, and this is where the agentic age needs a new operating system.\n \n \n DHH \n (01:58:24) \n When you can vibe code whatever app comes to your mind, you should be able to vibe code your operating system. You should be able to change anything, how it looks, how it works, what panels are there, what are not there, and that requires Linux. As simple as that. There are no\u2026 None of the other two operating systems can deliver this.\n \n \n Lex Fridman \n (01:58:41) \n But can you speak to the, I think somebody asked you, like, \u201cWho the hell are you to think you could do better than Ubuntu?\u201d For example. Or like, \u201cWhy are you doing it when Ubuntu is already pretty good?\u201d And your answer was, \u201cI thought I could do better.\u201d So can you speak to, this was on X, can you speak to, like, where the craziness comes from, thinking that you as one person or maybe with Ryan, you could start building something that\u2019s going to be better than Ubuntu or Arch or any of this?\n \n \n DHH \n (01:59:12) \n Well, I\u2019ll say that it didn\u2019t start quite as grandiose. When I got going with my Linux journey, I actually just started with Ubuntu, and I just built on top of it. Here\u2019s a system that already does a lot of the things I wanted to do, and at least it\u2019s Linux. So can I just build on top of it and make it more modern, make it actually look good, and have the programs I wanna use? I got, like, some version of that, and then I discovered that there\u2019s seven layers deeper you could go, and as, the closer you get to the metal, the closer you get to the kernel, the closer you get to the individual packages and so forth, the more freedom you discover.\n \n \n DHH \n (01:59:47) \n So as I discovered more and more freedom, I got more and more ambitious, and I realized, oh, surprise, surprise, I have strong opinions about how computers should work, how they should look, what programs should be installed, and whether they should take 42 minutes to install or 45 seconds. That\u2019s the current world record, by the way. So if anyone installs Omarchy Quattro after this and can beat 45 seconds, take a screenshot, send it to me, and you\u2019ll be the new holder of the world record.\n \n \n Lex Fridman \n (02:00:16) \n When\u2014how long ago did you cross the 60-second mark?\n \n \n DHH \n (02:00:19) \n It\u2019s funny because it\u2019s like that the old limit on, what was it? It was a four-minute mile or something- \u2026 like that in 1952- \u2026 that people thought was impossible for humans to beat. Then one guy beats it, and literally within the next few months, three more guys do, right? So we had this goal. I originally had the goal, it was two minutes.\n \n \n DHH \n (02:00:40) \n And then we beat two minutes, and I\u2019m like, \u201cWell, if we beat two minutes, why can\u2019t we beat one minute?\u201d And I remember thinking like, \u201cThat\u2019d be crazy. You can\u2019t install a modern operating system in one minute when they\u2019re all\u2026\u201d If the other ones are taking 42 minutes, right? Like Apple, full of a lot of smart people, like, it took 42 minutes, so that\u2019s gotta be the bar. No, no, it\u2019s not the bar because there\u2019s no speed limit. Like, none of the things you take for granted are fixed pillars of reality in almost any case. Until you get down to the seven gigabytes per second that drive is able to transfer, you\u2019ve not reached the limit yet, so you should just keep pushing. And of course, then it becomes a game in itself and a satisfaction itself.\n \n \n DHH \n (02:01:22) \n Every second I could shave off was an excitement. And when we broke the one-minute barrier, which was really only broken, I think, three weeks ago or something like that, maybe even two weeks ago, I just got crazy. And then I got a swarm of agents to run all these auto research loops of trying all these different theories on like, oh, if you change this thing, if you change that thing, what if you cut this out? What if you do things in parallel? What if while you were typing in your username, we start preloading packages into memory system, we can fire them in faster. And it becomes a fun game. And I think that\u2019s how you should think about product development. It should be fun. You should, it should be frivolous.\n \n \n DHH \n (02:01:57) \n It should be trying to overshoot the target and making it so much better than what it, quote unquote, needs to be.\n \n \n Lex Fridman \n (02:02:04) \n Especially now because you can move so quickly and you can try so many different approaches because of the agents.\n \n \n DHH \n (02:02:09) \n Yes, that just makes it more satisfying. But I think this was always true. If you look at the greatest Mercedes-Benz of all time, I think it was called the W126. I saw this wonderful documentary on it. It was steered by this guy who for years had worked in Mercedes\u2019 department for safety. And suddenly he gets put in charge for the whole S-Class project, and he just puts in all of it. This is where you got the auto-retending seat belts and washers on the headlights and everything. This car was so overbuilt for the market.\n \n \n DHH \n (02:02:42) \n It was unbelievable. But you look at that and go like, \u201cWell, I want that.\u201d In just the same way that you look at a mechanical watch and go, \u201cIt could go down to 100 meters?\u201d Or, or I think Rolex has one that can go down to 4,000 meters, the, the deep ocean, the deep diver ocean- \u2026 or whatever it\u2019s called. No one outside of three people in the world, whoever buy that watch, will ever put it to the test. But don\u2019t you wanna be the kind of person who is a patron of people who wanna push the human race that much forward? I want the fastest car breaking the speed limits. I want the diver watch that can go down the Mariana Trench. I want the operating system that can install in less than 60 seconds.\n \n \n Lex Fridman \n (02:03:26) \n So from the individual level, I admire those people. I think everybody should strive for that. But also the consequences of that we should mention, like the pursuit of excellence, even when the metrics seem to be frivolous, the maybe unexpected consequences of that is other kinds of excellences, discoveries, innovations, and all that kind of stuff that has nothing to do with the speed.\n \n \n \n \n Lex Fridman \n (02:03:49) \n It\u2019s discovery. It\u2019s nice to have the metric that you\u2019re chasing be the catalyst, the engine that drives innovation, because you\u2019ll just discover. It\u2019s the same reason you go to the moon, you go to Mars. It\u2019s like, why? Who cares? Well, actually, the real thing is you\u2019ll discover a lot about chemical propulsion. You\u2019ll discover a lot about how humans can live for a long time in space. Maybe you\u2019ll discover whole new ways of doing solar energy in space or data centers in space or whatever the hell.\n \n \n \n \n Lex Fridman \n (02:04:22) \n All kinds of\u2026 Maybe you\u2019ll meet-\n \n \n \n \n Lex Fridman \n (02:04:23) \n \u2026 aliens, but you never know. But setting a goal and chasing it, especially when it\u2019s numerical, is really nice.\n \n \n DHH \n (02:04:32) \n And setting a grand goal is even better. When you refuse to just do what\u2019s practical and you strive for what\u2019s slightly beyond plausible, that\u2019s when the magic happens. And we should be going all in on that. This is one of the reasons\u2026 And in fact, to some degree, this specific goal was motivated by seeing others do this. I mean, Elon is the singular individual in this conversation. The guy is so out there in his goals. They seem so preposterous at first glance. I\u2019ve certainly considered them preposterous on many glances. And then you realize that they come true sometimes. Maybe not always on schedule, and so what? So what? So we didn\u2019t get the self-driving car on schedule. I gave Elon a lot of shit for that back in \u201917.\n \n \n DHH \n (02:05:29) \n But now it\u2019s here, and who would have gotten us here except for the person who thought it was possible in \u201917, who is so impatient with reality that the sheer force is just gonna propel that reality to the place he wants it to be. I think that\u2019s deeply inspiring, and we need these kinds of goalposts to be moved. We need someone to run the four-minute mile, so the rest of us realize, \u201cWell, okay, but I guess I can run it quicker.\u201d That\u2019s what I try to strive for, and I really strive for with Omarchy because it\u2019s just more fun to live that way, to have high aspirations and ambitions. And I will say I\u2019ve actually changed my mind on that a little bit. I spent quite a lot of time speaking the other book that there are a lot of people who don\u2019t actually need those ambitions.\n \n \n DHH \n (02:06:22) \n And I also\u2013 That\u2019s also true. We don\u2019t want 8 billion Elons running around the world. Would be rather crazy, I think. But I\u2019ve come to realize that for entrepreneurs, it\u2019s healthy to have suitable chunky goals. The idea of just running the Italian restaurant can\u2019t just be that. It should be maybe I just run the one Italian restaurant, but I make the best damn pizza possible. I make the most delicious Parmesan. We have to strive for excellence to muster the motivation to keep going.\n \n \n Lex Fridman \n (02:07:00) \n And yeah, to dare to think big like you could build a Linux-\n \n \n Voice prompting vs typing\n \n \n \n Lex Fridman \n (02:07:05) \n \u2026 distribution. Is there some stuff you remember about what it took to get us to be so fast? You mentioned a few things that were, the agents were discovering.\n \n \n DHH \n (02:07:15) \n One of the things was to treat the lag of human input as an opportunity to preload. So there\u2019s five questions you answer when you set up an Omarchy machine. So it\u2019s doing stuff in the background. We\u2019re doing stuff in the background. And that was one of the things I didn\u2019t even consider, which is, I mean, the oldest trick in the book. All sorts of video games for a long time have done this, where you let the player interact with something, and then you do preloading and so forth in the background. So it\u2019s an age-old technique, but I just hadn\u2019t thought about it for an installer.\n \n \n DHH \n (02:07:45) \n I thought, first you ask the user some questions about their password and their username and their time zone and so forth, and then you do the work, when, in fact, you could preload that. Now, the other things were just a bunch of other tweaks to how the installer itself was doing things in a certain order. Some of them was also just shrinking it. The last version of Omarchy was 7.5 gigabytes, and a lot of the time we\u2019re down to now is literally decompressing compressed package files. That\u2019s the bulk of what the- is on the ISO, and that\u2019s the bulk of the install time. So by shrinking it down to, I think we\u2019re at 5.85 gigabytes now, a pretty chunky reduction, it almost literally translated one-to-one, and I did some really interesting things there.\n \n \n DHH \n (02:08:30) \n So Omarchy uses the JetBrains font, which is an awesome font, beautiful font, and actually a font I didn\u2019t like at first glance, but then it was the only font I could get to look perfect in Mitchell Hashimoto\u2019s Ghostty. Ghostty, for whatever reason, did not wanna render my beloved Bitstream Vera Sans just the way I was used to it. Mm-hmm. But it rendered this JetBrains font perfectly, so I switched over. But it\u2026 The standard package for JetBrains font on Arch is 200 megabytes because it includes all these variations and all these subtypes and whatever. And I realized, like, we\u2019re not using any of that. We\u2019re using the monospace version in this one Nerd Font patched edition. Do you know how big that is? I think it\u2019s 16 megabytes. Hmm.\n \n \n DHH \n (02:09:21) \n So I simply just came up with a new package, the slim version- Mm-hmm \u2026 of the JetBrains package. Nice. And there, right there, I saved 180 megabytes. I just went through and did that a bunch of times. We did that with the NVIDIA drivers too. So the NVIDIA drivers, when they came straight off the, I think the Arch repo, were not compressed with the most extreme form of compression, which I think called ZSTD, which is a really slow form of compression. It\u2019s slow to create the archive, but if you\u2019re building a package- Hmm \u2026 you have all the time in the world if that means you can then shrink the ISO. So on just the two NVIDIA packages, I think we saved 200 megabytes. Mm-hmm. So what I loved about this was I imagined myself being a car designer at McLaren.\n \n \n DHH \n (02:10:06) \n So McLaren makes some of the lightest super sports cars in the world right now, and they are absolutely obsessed with shaving every damn last gram off their cars. That\u2019s why even a new 750S McLaren is just way lighter than the competition. They use a carbon fiber monocoque, and that helps them over like Ferrari, for example, that uses aluminum. But they also just have this obsession with weight, and I would see this interview with the McLaren folks talking to some journalist just going over, \u201cYeah, we saved 370 grams off that one.\u201d And I just went, \u201cWait, what? There\u2019s someone on a car that weighs maybe 1,040 kilos or 1,380 or something. They\u2019re worrying about 370 grams?\u201d Yeah. What obsessed maniacs. Yeah. I love them. Yeah. I want that in my life.\n \n \n DHH \n (02:11:02) \n I want a McLaren just for that reason. So I thought in that moment as I was shaving individual megabytes off the packages, man, I\u2019m kinda like a McLaren car developer here, even though I mean, we haven\u2019t even completed that work. The package could shrink even more. But this is actually this other conflict that we\u2019ve had with Omarchy and the traditional Linux community is bloat. This idea that it\u2019s actually not kosher in certain circles to ship with pre-installed software because who are you to make decisions about what programs people should use? And I\u2019m like, \u201cDude, it\u2019s in the fucking name. It\u2019s called Omarchy because the Oma part is short for omakase, which literally means chef\u2019s choice. I\u2019m the chef.\n \n \n DHH \n (02:11:47) \n I\u2019m making the choices, and this is the collection of applications I think is awesome,\u201d which by the way has a video editor. It ships with OBS to do your recording. It\u2019s just in there. Yeah. OBS is pre-installed. Nice. Kdenlive is a video editor, timeline editor, which is great. I edit all my videos with that. And then I made my own clip editor. You talked about the thing with syncing up. Yeah. I had this thing with clips, so I would do a podcast or whatever. I need a clip for X. And it\u2019s kind of a hassle to use Kdenlive or other timeline editors, so I made Omacut. Omacut is included too. Omacut, cool. It\u2019s just for making clips, and you can drive the whole thing with keyboard. So you can go to just the right time mark.\n \n \n DHH \n (02:12:25) \n and then you can hit Control + Space. It moves the clip line to that line. Then you go to the end of the clip, and you hit Alt + Space, and it moves the end line. Nice. And then you hit Control + S, and it saves. Nice. I can make clips so damn fast now with Omacut. It\u2019s amazing. But it has all the software on it. It has Neovim on it. It has Herdr on it. It has Tmux. It has a terminal. It has a bunch of themes. It has a bunch of background images, the best ones I could pick out. That takes up some space. It should because this is not supposed to be this just barren landscape where you have to reconstruct everything. This is supposed to be a productive system the minute you unpack it. What is it built in?\n \n \n DHH \n (02:13:05) \n So it\u2019s actually mostly built in Bash for the Linux configuration part. Right. Bash is a great language for that. It is made for that really, for system administration. It also has its limitations. You should not build everything in Bash.\n \n \n Lex Fridman \n (02:13:21) \n How are agents with generating Bash? Pretty good?\n \n \n \n \n \n \n DHH \n (02:13:24) \n Except for one thing, and I have to smack the agents over the back of the head every single time I catch it, and remind it to look at the agent\u2019s MD file, because I will have an instruction in there not to make early exits. The agents love to have preconditions instead of fully expanded conditionals. I like something that says, \u201cIf this, then do that.\u201d Agents like to have precondition or exit, another precondition or exit.\n \n \n DHH \n (02:13:53) \n And then it falls down to the final thing it actually wants to do. I hate that style. So except for that, it\u2019s gotten shockingly, dramatically, awesomely good, to the point where when we talked last time, we talked about how I would ask the agent for a Bash thing, and then I would type it in myself, and I kinda wouldn\u2019t remember. And then I took the effort to make sure that all that competence wasn\u2019t draining out of my fingers, so I actually learned how to write Bash myself, and now I\u2019m not writing any Bash myself. I have not written any Bash myself for probably a couple months, because the agents have just gotten so good, that if you give them the few pointers on style, and occasionally put pushback on complexity\u2026 This is\u2026 I tweeted about this a couple of days ago.\n \n \n DHH \n (02:14:32) \n It blows my mind how often an agent can do something, say it\u2019s done, have the work reviewed by another agent- \u2026 they say it\u2019s done, and then I go like, \u201cEh, looks a little too complicated for me.\u201d And then it\u2019ll go like, \u201cOh, yeah, you\u2019re right. You\u2019re right. I totally overcomplicated this,\u201d and it\u2019ll cut it in half. I\u2019m like, \u201cCouldn\u2019t it just do this, like, from the outset?\u201d And then someone else reminded me, humans are the same way. If you work on something and someone else reviews it and they just are like, \u201cEh, it looks a little complicated,\u201d quite often you\u2019re able to go like, \u201cAh, you\u2019re right.\u201d\n \n \n Lex Fridman \n (02:15:05) \n See, that just\u2026 Those little indications to me show that the programmer is still needed, that intuition, \u2019cause I think, what did you say? I have a sense of the landscape of the code.\n \n \n DHH \n (02:15:17) \n When I build something like Omarchy, I do look at the code. I look at the shape.\n \n \n \n \n DHH \n (02:15:23) \n And I look at the proportions.\n \n \n \n \n DHH \n (02:15:24) \n Like, I don\u2019t have to chisel everything myself anymore. This is, again, when I\u2019m flattering myself for\u2026 Not one moment I\u2019m a McLaren auto engineer, and the next moment I\u2019m a da Vinci, working with his whole studio of students- \u2026 who are doing all the chiseling on the, the fine marble, right? And he goes like, \u201cAh, no, the proportions are not quite right.\u201d That\u2019s how it feels working with agents now, that I can give feedback like an editor.\n \n \n DHH \n (02:15:50) \n That, do you know what? I\u2019m not gonna write everything myself, but I can tell you when the argument doesn\u2019t land. I can tell you when the shape isn\u2019t proportional to the problem, or when it just feels like it\u2019s too complicated. And shockingly often, it just works. It\u2019s like the early days, where the joke was, \u201cDon\u2019t make mistakes.\u201d I don\u2019t think you have to tell the agents that anymore, because now the harnesses are so good, they run automated testing. They get that feedback. But you actually do have to tell them, \u201cMake it simpler.\u201d\n \n \n Lex Fridman \n (02:16:19) \n Do you use voice at all for the interaction? Or are you\u2026 How much are you typing? How much are you-\n \n \n DHH \n (02:16:23) \n I type the whole thing. I actually, in Omarchy\u2019s built-in Voxtype- \u2026 Audio transcription, and it\u2019s actually really good. I just find that I don\u2019t think that way, and it surprised myself that I don\u2019t like to talk to the computer. I like when I\u2019m in the car, but when I\u2019m in front of the keyboard, I just wanna type. I also just love typing. Are you using the voice?\n \n \n Lex Fridman \n (02:16:46) \n I\u2019m an introvert who doesn\u2019t like social interaction- \u2026 or speaking. And-\n \n \n DHH \n (02:16:50) \n I think that\u2019s part of mine too.\n \n \n Lex Fridman \n (02:16:53) \n But, like, for me, for some reason, voice is really powerful. I want to give this\u2026 So I run with this thing. It\u2019s called Plaud. So I just hook it up on my, you know on my T-shirt here.\n \n \n Lex Fridman \n (02:17:05) \n You press a button, it starts recording. And this is actually more, in terms of flow, much better experience than behind a computer. And I save, well, usually it\u2019s agent generated problems that I wanna really talk through. And so I write, I speak a long prompt, like, and a lot of it, it\u2019s like stream of consciousness reasoning. So I\u2019ll sometimes just change my mind about a design, but all of that is in there, and then I have an LLM that processes that. With Plaud, there\u2019s a, obviously an API. You can grab it, transcribe it, clean up the transcript, \u2019cause you want it to be\u2026 \u2018Cause I\u2019m referring to files. I\u2019m referring to technical concepts, like files and functions- \u2026 and so on, that you wanna make sure the transcription, the-\n \n \n Lex Fridman \n (02:17:56) \n \u2026 the speech-to-text gets correctly. So you wanna have a dictionary. You want it to be aware of the code base when you process the files. And then you have this gigantic thing that\u2019s often 10, 15, 20 minute me discussing the full design, and I find that to be really, really powerful. Because it doesn\u2019t have the problem of overspecification because of how much stream of consciousness thinking the system has about what you\u2019re imagining. And this is really good for the early steps of a design. Like, if I wanted to build Omarchy from scratch, the first stages, I talk through. It could be an hour. There\u2019s been prompts that I had for an hour, and I\u2019m just running and-\n \n \n \n \n Lex Fridman \n (02:18:40) \n \u2026 and speaking. I highly recommend people try a voice-based-\n \n \n \n \n Lex Fridman \n (02:18:48) \n \u2026 speaking for prolonged periods of time.\n \n \n \n \n Lex Fridman \n (02:18:51) \n As opposed to very specific\u2026\n \n \n DHH \n (02:18:52) \n \u2026 I haven\u2019t, yeah, I haven\u2019t used it like that, and maybe that\u2019s how I\u2019m using it wrong. I try to just type or speak two sentences, and then I get a little frustrated that I could\u2019ve typed that just as fast, maybe. But I have not tried this whole thing of just talking continuously for 20 minutes.\n \n \n Lex Fridman \n (02:19:09) \n And especially you because you speak very fast.\n \n \n DHH \n (02:19:12) \n Yeah. No, I could see it. I think because the reason I\u2019ve been trying to do it for things like replying to emails, and it got to a point where I thought it could help. And it sort of did, but then it would make a small mistake, and I\u2019d have to correct the mistake. But if you\u2019re doing stream of consciousness to an LLM who doesn\u2019t care about all this other stuff anyway\u2014 \u2026 that probably works way better.\n \n \n Lex Fridman \n (02:19:34) \n There\u2019s two things you have to do. One, okay, you have the raw audio, transcribes it. I use ElevenLabs for transcription. They\u2019re probably the best speech-to-text model. There\u2019s still mistakes there, so you need to have a prompt that has access to a dictionary and to a lot of\u2014\n \n \n DHH \n (02:19:50) \n How do you build the dictionary?\n \n \n Lex Fridman \n (02:19:52) \n First of all, there\u2019s a dictionary of terms that are relevant to you. Then have it aware of\u2026 So you have an agent that looks through, when you\u2019re talking about code, looks through your entire code base for key terms that you use often. Also, when we\u2019re talking about messaging, Gmail, and WhatsApp, and so on, it\u2019s aware of all the people and common terms that you get wrong.\n \n \n DHH \n (02:20:11) \n You\u2019ve really gone deep in this.\n \n \n Lex Fridman \n (02:20:13) \n That, to me, is a problem that needs to be solved-\n \n \n \n \n Lex Fridman \n (02:20:15) \n \u2026 because one of the issues you probably have with speech-to-text is that it makes mistakes sometimes.\n \n \n DHH \n (02:20:22) \n Correct. Then I can\u2019t use it for dictating an email.\n \n \n Lex Fridman \n (02:20:25) \n Right. If it never makes mistakes, you-\n \n \n DHH \n (02:20:28) \n Then it would be more appealing.\n \n \n Lex Fridman \n (02:20:29) \n Yeah, and you are okay being not perfectly real time, actually. You\u2019re okay waiting five seconds-\n \n \n \n \n Lex Fridman \n (02:20:35) \n \u2026 potentially. I don\u2019t know what your use case is. I\u2019m okay waiting five seconds- \u2026 10 seconds for it to process everything.\n \n \n DHH \n (02:20:43) \n You gotta vibe code this into an app that does all of this such that-\n \n \n \n \n \n \n Lex Fridman \n (02:20:47) \n I think I mentioned to you offline that it\u2019s very surprising to me that translation is not done perfectly.\n \n \n \n \n Lex Fridman \n (02:20:58) \n Speech-to-speech translation.\n \n \n DHH \n (02:20:59) \n It seems like you\u2019ve gotten much closer to the best of what\u2019s possible by putting all these things together.\n \n \n Best AI coding models\n \n Lex Fridman \n (02:21:05) \n But, you know, when you\u2019re shipping that kind of stuff, you do have to start thinking about latency.\n \n \n DHH \n (02:21:11) \n Don\u2019t let the perfect be the enemy of good. I mean, if you\u2019re suddenly enabling something people wouldn\u2019t be doing before at all, it\u2019s okay to take five seconds longer. I think this is one of the areas that Steve Jobs was so good, that he realized, like, the first iPhone was terrible in all sorts of ways, right? Absolutely horrendously slow data connection, very underpowered, all the things. But the product was so compelling that it just didn\u2019t matter. You\u2019d be shipping too late if you waited until everything was just right, just perfect. If you\u2019re solving a problem that people currently don\u2019t have a solution for, they\u2019re willing to give it five seconds.\n \n \n DHH \n (02:21:49) \n If you\u2019re in a competitive market, yeah, okay, it\u2019s different. But that means the problem\u2019s already solved, and you don\u2019t need to solve it.\n \n \n Lex Fridman \n (02:21:54) \n Mm-hmm. Well, I do think there\u2019s a few\u2026 So, like, Whisper flow is really good. A few people have stepped up that speech-to-text thing.\n \n \n DHH \n (02:22:02) \n Yeah. Yeah, we\u2019re using VoxType.\n \n \n \n \n \n \n Lex Fridman \n (02:22:05) \n VoxType is open source?\n \n \n DHH \n (02:22:07) \n It is. It\u2019s just using one of the open models. I forget which parrot model it\u2019s using or something like that. It works quite well for short commands, so I have used it for that. I think you just hold down F9- \u2026 and it starts a dictation.\n \n \n Lex Fridman \n (02:22:21) \n Oh, so it\u2019s already in there?\n \n \n DHH \n (02:22:23) \n You just have to get online. It\u2019ll offer you to install- \u2026 the VoxType package. Because the VoxType package includes a model that\u2019s 150 megabytes, and I was like, \u201cAh.\u201d \u201cI don\u2019t know if I can carry that.\u201d So- \u2026 we don\u2019t have that in there by default, but we have the option set up. And I actually just yesterday was thinking we really need to combine all of these things. So Omarchy Quattro is truly amazing as a malleable operating system where you can add functionality to it just by talking to your agent.\n \n \n DHH \n (02:22:56) \n But that\u2019s exactly what we should be doing. There\u2019s a lot of people who would find it very natural just to talk to their computer and say, \u201cHey, can you make me a stock panel? I wanna track Apple and Dell,\u201d and just see the agent go off through, entirely through voice, both in terms of the input and in terms of the output, and then your operating system just changes. This was the vision of Iron Man and Jarvis and-\n \n \n DHH \n (02:23:23) \n \u2026 bespoke applications that just appear magically. I\u2019m thinking the area of AI I find truly intriguing are these live gaming models, I don\u2019t know if you\u2019ve seen that, where the AI\u2019s inventing literally the next frame, but you can actually play the game. And I think if they\u2019re able to do that, why can\u2019t we do that with our operating system? Why can\u2019t we just talk to it and ask it to be a different way or change, and it\u2019ll all just change? I think this, by the way, is one of the other breakthroughs that is going to lead to the total domination of Linux. The agents have taken all the hardship out of diagnosing Linux systems and turned the fact that Linux produces these overly specific, totally arcane error messages into its greatest advantage.\n \n \n DHH \n (02:24:12) \n When Linux has an issue, the agent can take that very specific error message that makes no sense to a normal human and correlate it with the fact that the agent was pre-trained on 40 million lines of Linux code, so it knows exactly where to look and dial it down. I have not had a single problem on my Linux machine since the beginning of this year that an agent could not diagnose. That was not true a year and a half ago. A year and a half ago, I was still searching on forums- \u2026 to find answers to esoteric questions.\n \n \n DHH \n (02:24:45) \n Now the agents know the source code of not just the Linux operating system, but every single piece of software I have on that box. The agent has access to the source code of all of it. In fact, this was one thing I built in just before we shipped. So once you set up your default agent- \u2026 Omarchy Quattro has a crash watcher. If any app on your machine crashes, it\u2019ll pop up a little thing, ask you whether you want your AI to diagnose the problem.\n \n \n Lex Fridman \n (02:25:13) \n Yeah. That\u2019s amazing.\n \n \n DHH \n (02:25:14) \n Unbelievable. I\u2019ve seen things where there\u2019s an error in some subsystem. Some\u2026 Not even Linux, just some app I installed. The agents start digging through the logs and look up systemd log. Then it checks out the damn source code. Of the application that crashed, pins down that it\u2019s in this Rust file line 472 that there\u2019s an unbounded, unwrapped variable that overflowed or whatever it is. Then offers you whether you wanna file a bug report with all this detail. I\u2019ll give you one amazing anecdote. So I was working with this guy JDX, who\u2019s working on mise. This, this package manager for fast-moving development tools. This is how we manage all the agent software and so forth. Normal package managers are not built to be updated seven times a day.\n \n \n DHH \n (02:26:04) \n And all of these agent harnesses are updated about seven times a day, so we needed an out-of-band package manager, and mise just turns out to be perfect for this. Anyway, it had a small bug. There was some race condition which agents are very good at finding because when you start running multiple agents at the same time, they will suss out all these race conditions you have in your underlying infrastructure that was never triggered by a human trying to do manual thing. So it finds this issue, right? I tell the agent, \u201cHey can you post this as a bug report to, to JDX\u2019s GitHub?\u201d And unfortunately, right before that, I had had to do a QA run on Omarchy itself with eight different agents. It had found 28 real issues-\n \n \n DHH \n (02:26:48) \n \u2026 that it needed to file. Well, it went to GitHub and tried to file all 28 issues at the same time, which it did in about I don\u2019t know, 12 seconds. GitHub, not unreasonably, marked that as probable spam and banned- \u2026 my Omarchy bot. And then I couldn\u2019t do that. So I was blocked from the bot having access to GitHub, so I just told it, \u201cHey, do you know what? Just email JDX. Here\u2019s his email address.\u201d I had already set it up with hey.com- \u2026 an email address. We have a CLI that\u2019s in beta right now.\n \n \n DHH \n (02:27:20) \n So I set it up with that so it can send to email. That\u2019s how it gives me reports about outstanding issues and, and PR. So it sends JDX this email about the bug it found, and he was like, \u201cWait, this is the first time I\u2019ve gotten a bug report in unreleased software.\u201d Because it had downloaded the source code to mise, saw that he had already fixed the bug, but that it would not fix the problem entirely in software they had not shipped yet. Pinpointed the problem, and he was like, \u201cDamn it- \u2026 I got a bug report before we even cut a release.\u201d\n \n \n \n \n \n \n \n \n DHH \n (02:27:54) \n \u2026 believable AGI levels of mind-blowing stuff.\n \n \n Lex Fridman \n (02:27:58) \n And that was actually, y- if you remember, that was, that was an open question whether AI systems could debug well. That, that was kinda the, the, the thought that, yeah, sure, they can write code, but can they find issues and-\n \n \n DHH \n (02:28:10) \n They are so incredibly good at this, and this is actually interesting because I was on the perhaps same side of that, a little skeptical about can it reason about all these things. And Mikhail, who\u2019s the CTO at Shopify, ran a scientific study on this. They had\u2026 And this was, I think, late last year or early this year, where he had agents go back through all the incidents, both outages and other problems that Shopify had had in production, trace that back to the PR that was merged, and find out whether PRs that had been reviewed by human or PRs that had been reviewed by agents were of higher quality. Well, lo and surprise-\n \n \n DHH \n (02:28:50) \n \u2026 the PRs that have been reviewed by agents caused all fewer issues in production. And this was with models we had six months ago. At this point, 100% in the majority of domains we work in today, agents are better at finding bugs.\n \n \n Lex Fridman \n (02:29:06) \n So you now run agents all day. What stands out to you as the better model? Who\u2019s currently winning? It seems to be changing constantly. Sol, Fable, Grok 4.6, Gemini, the Chinese OpenWay models.\n \n \n DHH \n (02:29:23) \n What\u2019s amazing to me is that you could rattle off so many different contenders. That this market is so wide open, that it does actually change back and forth, that we have real competition, that there are so many labs that are able to get either to the frontier or close to it. That, by the way, is remarkable. I still don\u2019t fully understand that. But to answer your question, the best model in general right now is Fable. The second-best model, in my opinion, is Opus 5. But the tier just below Opus 5, and it\u2019s not even that they\u2019re always below, sometimes they\u2019re ahead. I\u2019ll get to that in a second. I would rank GPT Sol as very good.\n \n \n DHH \n (02:30:12) \n Grok 4.6 I just started testing a few days ago. I had this wonderful test that I\u2019ve set up by accident where I translated this Python library into Rust. The screensaver you just saw with all the cool animation?\n \n \n \n \n DHH \n (02:30:28) \n That\u2019s powered by a Python library called Terminal Text Effects. Really cool library. We\u2019ve been using it since the first day of Omarchy. The problem with that is it\u2019s written in Python, so when it starts up, especially on a laptop, and it runs in Python, it uses all of your CPU to do these effects, and therefore, it means it uses about 30 watts of energy, and it spins up your fans, and it drains your battery.\n \n \n DHH \n (02:30:52) \n Doesn\u2019t really matter on a local computer, but it does matter on a laptop. So I thought, \u201cDo you know what? This sounds like a problem for Rust.\u201d So first I gave Fable the challenge, and all I told it was, \u201cHere\u2019s the source code for the Python library,\u201d this TTE library that had a bunch of dependencies and so forth. \u201cI want a Rust version of this with no dependencies, a single executable.\u201d Like, that\u2019s what Rust does. So basically, I want it in Rust. I want it to be pixel perfect, frame by frame, do a full analysis, don\u2019t stop until you\u2019re finished.\n \n \n \n \n \n \n \n \n DHH \n (02:31:29) \n I kid you not, in just under 45 minutes, it was like Am I heard all?\n \n \n \n \n DHH \n (02:31:36) \n I\u2019m finished. I\u2019ve checked everything. I have reduced the startup time from 86 milliseconds to two milliseconds. I have sped up the execution time by 9.6 times, I believe it was. The executable is three megabytes. Do you wanna run it?\n \n \n \n \n \n \n \n \n DHH \n (02:32:02) \n \u2026 and I don\u2019t know why I\u2019m surprised, because this translation job is something we\u2019ve known for a while that AI is pretty good at, but it was still staggering to me that I could one-shot a full translation of a Python library I had been using for a year, that others had been using for much longer, and turn it into a Rust executable without knowing any Rust, without looking at the Rust code at all, and produce this executable that I then told the agent right after, \u201cThis is great. Ship it.\u201d It packaged it up as a new package. It told me, \u201cWhat do you want it to call?\u201d \u201cUh, let\u2019s call it TTFX. Let\u2019s create a new Git repo.\u201d It sets up the Git repo. \u201cLet\u2019s create a new package, build package for our build system.\u201d It puts that up. \u201cLet\u2019s push it out.\n \n \n DHH \n (02:32:49) \n Let\u2019s open the pull request to the Omarchy itself so we switch around from TTE, the Python implementation, to TTFX.\u201d It does all of it, and I\u2019m just sitting there. And again, I\u2019m already at this point fully delirious with agent acceleration, and I still had to lean back and I go like, \u201cThis is AGI, isn\u2019t it? This is what AGI looks like. If we have\u2026 If these moments are just what it is all the time, this is AGI.\u201d\n \n \n Lex Fridman \n (02:33:15) \n Now, as I understand, you also repeated the same experiment not just with Fable.\n \n \n DHH \n (02:33:21) \n I did it on all the models. First, I did it\u2026 Actually, funny thing. So I ran out of Fable tokens about two-thirds through, and it just automatically switched over to Opus 5 and kept going and finished the job. Part of the reason why I think it was able to do that was that the first thing Fable did was create a plan, and it was a really detailed plan. I think it had eight separate steps of, \u201cHere you do it, and here, how you analyze it, and here you run the effects,\u201d and so on. I didn\u2019t review the plan at all. I didn\u2019t change the plan. It just made the plan so Opus could take over. And then I thought, \u201cWell, if Opus can finish the job, maybe some of the other agents could finish it, too.\u201d So the first thing I did was I gave it to\u2026 I think I gave it to Sol.\n \n \n DHH \n (02:34:05) \n And Sol finished the job, too. It took twice as long, so it took, I think, about an hour and a half. But here\u2019s the kicker. The per token cost, I didn\u2019t pay per token. I have a max subscription to Claude, right? So it did it within the subscription. It used all my tokens, so I had to switch over to Opus 5. But if I had paid per token, it would\u2019ve been 550 bucks, I think, to do the whole thing. And I thought for a second, \u201cHoly shit. What a steal.\u201d If I personally had to learn Rust well enough to be able to do this translation, I\u2019m looking at a nine-month job here. I can pay 500 bucks to have this translation happen, and suddenly I get a 10x execution speed up. This is\u2026 I would totally pay 500 bucks for this.\n \n \n DHH \n (02:34:48) \n But competition. So I give it to Sol. Same plan. To be fair, I didn\u2019t ask Sol to do the plan. I just took the Fable plan, gave it to Sol. Sol, in an hour and a half, and I think $46 worth of tokens, repeated the task. Did the same thing. And I thought, \u201cWell, blimey, that\u2019s amazing.\u201d Then I got greedy. So I asked GPT Luna, which is this crazy cheap model that OpenAI has as well, \u201cCan you do it?\u201d Absolutely not. First of all, it didn\u2019t even wanna start the task. I think something happened\u2014I don\u2019t know, I think it was in the spring\u2014where we didn\u2019t need these slash goal things anymore. The agents could just automatically keep going in a loop if you told it not to stop-\n \n \n DHH \n (02:35:44) \n \u2026 and so forth. So Sol could do that. Fable could do that. But Luna couldn\u2019t. So I had\u2026 I think I did 12 prompts. Kept telling it to do it, and eventually I sort of got it started. The first thing it did was to cheat. So the first thing, it looked outside its own directory, saw that there was already another implementation, and just did a short wrapper around that and said, \u201cI\u2019m done.\u201d Hilarious. But it couldn\u2019t finish because it made just a couple things. But I mean, okay, so it can\u2019t do that. Then I gave it to Grok-\n \n \n \n \n DHH \n (02:36:15) \n \u2026 46. And I had used Grok 45 a little bit, and I thought like, \u201cAh, I mean, it\u2019s cool that there\u2019s others trying, but, like, I\u2019m not gonna use it,\u201d because it felt quite far behind. Grok 46 fucking completes the task. 10x speed up, same size executable. $55 Worth of per token cost, I think it was. Absolutely unbelievable. Then I repeated, too, with Kimi K3- \u2026 which took forever. I forget how long Kimi actually spent on it. And then I also did it with DeepSeek V4 Flash first. And Flash failed the same way that Luna did. It couldn\u2019t do it. And then I did it with Pro, and Pro also completed the task. It took 2 hours 45, $23. So here we are, right? Like, Fable, clearly the best. It was the fastest. It was the one that wrote the plan, but 550 bucks, and the output the same.\n \n \n DHH \n (02:37:12) \n The others, Sol, Grok, about the same, 1/10 the cost. DeepSeek, 1/20 the cost, but you have to wait a little longer. Absolutely gobsmackingly incredible. And now, by the way, so I had Fable finish the first job, and then Opus finished it. That was the first one shot, right? It\u2019s 10 times faster. I did two auto research runs, which isn\u2019t even auto research anymore. You don\u2019t have to do the slash. You just tell it to keep going until you tell it to stop. It ended up\u2026 I think we ended up with a 46-time execution improvement over the original.\n \n \n \n \n \n \n Lex Fridman \n (02:37:52) \n It\u2019s just incredible.\n \n \n Best AI coding harnesses\n \n Lex Fridman \n (02:37:55) \n I do, you know, I just don\u2019t think there\u2019s anything that compares to Fable in terms of planning. So I usually do Fable for planning and for reviewing and then something else for implementation, like Opus 5. And it\u2019s really nice. It\u2019s a really nice setup that allows you to not run out of tokens too much.\n \n \n DHH \n (02:38:14) \n What I found, it\u2019s really interesting because Fable is, in my opinion, the best model right now, but it also makes mistakes. And the best way to get the best software, I would actually rather have two differently sourced sort of\u2026 I mean, they\u2019re not mid-tier, they\u2019re all frontier. But have, let\u2019s say Opus 5 and Codex- \u2026 and have one check the other\u2019s job. This is my standard operating procedure now. I\u2019ll have a Opus or Fable do the work, and then I always end it, review with Codex xHigh.\n \n \n DHH \n (02:38:47) \n And I\u2019ve also started using Grok just to test it out, and it\u2019s also quite good. And it keeps finding stuff. And then that\u2019s my workflow when I\u2019m having my agents on my own machine do it, and then I push to GitHub. And then Copilot, I kid you not, has actually gotten good. Copilot keeps finding stuff that\u2019s legitimately broken, which is also incredible acceleration because the first version of Copilot that started doing this was literally retarded. It would just constantly flag things that were nonsense. It would constantly flag the same problem over and over again as you would push. It was really annoying, so I think a lot of people actually ended up turning that off. And if they did, they should turn it back on because it\u2019s actually quite good.\n \n \n DHH \n (02:39:24) \n Keeps finding things. And if you then take that, and we shouldn\u2019t be surprised. Why are we surprised? Even if you\u2019re a good programmer, if you finish a job and you ask your also very good peer to review it, you\u2019re gonna end up with better code. Of course you\u2019re gonna end up with better code. So build that into your process. Pick one of the agents to drive with. I\u2019ve mainly been driving with Claude, which is actually interesting because I have some other reservations about Anthropic. But the reason I\u2019m sticking with Claude is, in my opinion, they actually have the best harness. And one of the reasons it\u2019s the best harness is it\u2019s multi-agent running. So if you wanna run multiple agents at the same time, you can do arrow left when you\u2019re inside a session, then it goes back to Agent View.\n \n \n DHH \n (02:40:06) \n And here in Agent View, you can pick up another agent. So if you wanna do this thing where you have multiple threads going on- \u2026 Claude Code is just the nicest setup. They also just, they keep being a little further ahead, which probably shouldn\u2019t be surprising. I mean, Boris, one of the guys that\u2019s working on that, he was the first one and basically came up with the thing. It\u2019s just interesting that that has been an enduring advantage. I\u2019ve also used OpenCode a lot. I think OpenCode is great. I use OpenCode mainly as my main harness for all the open models, so Kimi K3 and so forth. And I inference not on the Chinese servers. So I inference on Fireworks-\n \n \n DHH \n (02:40:43) \n \u2026 which is a really nice service where you can just pay by the token, and if you\u2019re using these open weight models, they\u2019re not that expensive by token. So that\u2019s a great way to do that. But I don\u2019t think I would be using\u2026 Well, maybe I would. But if\u2026 Using Claude with a subscription does feel like a bargain, like a crazy bargain.\n \n \n Lex Fridman \n (02:41:03) \n Yeah, I knew. And I actually have four.\n \n \n DHH \n (02:41:06) \n I just signed up for my second one this morning- \u2026 because I got up at 4:00 with jet lag, and I started working with the agents right away, and now I\u2019m three days away from limits resetting on Fable, and I ran out of tokens.\n \n \n Lex Fridman \n (02:41:19) \n So, and they should really-\n \n \n DHH \n (02:41:22) \n I mean, I don\u2019t\u2026 Why is that so complicated? Can\u2019t you just stack one subscription? Why do I need to log in multiple times?\n \n \n Lex Fridman \n (02:41:26) \n And obviously you\u2019ll have to develop some kind of tooling-\n \n \n \n \n Lex Fridman \n (02:41:31) \n \u2026 to move from one- \u2026 to the other. And then-\n \n \n DHH \n (02:41:33) \n Oh, we\u2019re building that into Omarchy, by the way. So the next version of Omarchy is gonna ship with multi-sub support. I wish that the labs would just make you buy a max 100 times.\n \n \n Lex Fridman \n (02:41:43) \n It\u2019s kind of actually fascinating though for me because there\u2019s\u2026 It\u2019s very challenging for me to use Grok 4.6 and then a Claude Code because Grok is so fast.\n \n \n DHH \n (02:41:57) \n Yes. That\u2019s actually the deal. The deal with Grok is that their fast mode is cheaper than the regular mode on the others. And when you\u2019ve used Grok 4.6 on fast, it\u2019s kind of addictive \u2019cause it almost gets you to this single thread mode- \u2026 where the agent is able to keep up with you.\n \n \n Lex Fridman \n (02:42:14) \n Yeah, but I don\u2019t know how\u2026 I can\u2019t\u2026 Once I get used to Claude Code I can\u2019t handle the speed of Grok \u2019cause I\u2019m used to now multitasking.\n \n \n \n \n Lex Fridman \n (02:42:24) \n And so now I\u2019m back in this-\n \n \n \n \n Lex Fridman \n (02:42:26) \n \u2026 anxious mode of, switching.\n \n \n \n \n Lex Fridman \n (02:42:27) \n \u2026 switching. So I mean, all of this is a learning process of what programming actually is supposed to look like.\n \n \n DHH \n (02:42:34) \n By the way, one of the reasons why I\u2019m not sure what the final form of these harnesses is gonna take, because one of the things we\u2019ve started experimenting with at Basecamp is that we put the agents inside of Basecamp, and we treat them as coworkers.\n \n \n Lex Fridman \n (02:42:45) \n Yeah. That\u2019s awesome.\n \n \n DHH \n (02:42:45) \n And then we assign them tasks-\n \n \n \n \n DHH \n (02:42:47) \n \u2026 inside of Basecamp, so they will do work on a to-do in Basecamp, or they\u2019ll work, do work on a card. And it turns out that a collaboration tool that\u2019s optimized for asynchronous communication is actually the right format versus these harnesses are a little more like chat. And chat, which was, I mean, the original format for these things, is not the right thing because you\u2019re sitting around waiting. It entices you to sit around and wait, versus when you\u2019re dealing with something like Basecamp and you can just give an agent a to-do item, you don\u2019t expect to hear back immediately. So I don\u2019t know what the final form is gonna take, but I\u2019ve been using all of the harnesses. So I use them inside of Herdr, and Herdr has panes.\n \n \n DHH \n (02:43:27) \n So oftentimes I\u2019ll run Claude up top, and then I\u2019ll run Codex down below, and then maybe I also have an OpenCode set up. But I will say just very recently, I found that the human in the loop is the limit here. So I\u2019ve started setting up more automated systems. I\u2019ve been building an Oma bot that can do more autonomous development on Omarchy that is just on a regular schedule process, all depending to do or PRs and issues, and then send me an email using the HEY CLI, and then I just get this email. Here\u2019s 12 PRs that are either ready to go or I think you should close. And then I just make the final determination there.\n \n \n Lex Fridman \n (02:44:06) \n Do you find the multitasks, the task-switching mentally exhausting?\n \n \n DHH \n (02:44:10) \n Yes, but in an exhilarating way. Like, one of the things I always loved about race cars was when I would stumble out of the car absolutely smashed and barely able to hold my head up, and I\u2019d lay down on the garage floor and just think, \u201cHoly fuck, I\u2019m alive.\u201d That\u2019s the kind of exhaustion, mental exhaustion I\u2019m feeling at moments with the agents right now.\n \n \n Lex Fridman \n (02:44:31) \n By the way, when you\u2019re racing, where, what\u2019s the source of the exhaustion? Like-\n \n \n DHH \n (02:44:34) \n It\u2019s physical. It\u2019s just\u2026 I mean, I bet it\u2019s the same thing with jujitsu, right? Like, it\u2019s actually satisfying to feel exhausted. It feels like you\u2019ve applied yourself. And I feel that with the mental exhaustion you can get from agents.\n \n \n Lex Fridman \n (02:44:48) \n Are you also mentally drained from racing?\n \n \n DHH \n (02:44:49) \n Yes. If you drive in difficult conditions, especially if you drive in the rain, where you\u2019re constantly managing the thing right at the knife\u2019s edge. But most of the time, I just enjoyed the physical exhaustion.\n \n \n Lex Fridman \n (02:45:01) \n And here, with the multitask switching, I do find that, like, I get really exhausted, and I think it\u2019s because I\u2019m thinking a lot. \u2018Cause, like, you\u2019re basically-\n \n \n \n \n Lex Fridman \n (02:45:14) \n \u2026there\u2019s no, like passive low energy thinking.\n \n \n \n \n \n \n \n \n \n \n \n \n Lex Fridman \n (02:45:21) \n Solve problem, next problem, solve problem, and you\u2019re really thinking.\n \n \n DHH \n (02:45:25) \n It\u2019s funny. It\u2019s actually very similar with racing. There\u2019s some tracks, like Le Mans, where you get time to relax. There\u2019s the Mulsanne straights where it\u2019s kinda long. You\u2019re going fast, but in a straight line, you can take a breath. And then there are other tracks where it\u2019s just coming full on all the time, and you don\u2019t have any moment to relax, and you stumble out of the car after an hour in each, and it\u2019s very different how exhausted you are. And this is exactly what I\u2019m finding with the agents too, that when you\u2019re running at max human capacity- \u2026 you\u2019re just constantly in a corner.\n \n \n Lex Fridman \n (02:45:57) \n Do you think people are in danger of burnout?\n \n \n DHH \n (02:46:01) \n I think the moment we\u2019re in right now is gonna pass. That this need to\u2026 It\u2019s not a babysitting motion, but constant interaction with the agents is gonna fade. I\u2019m, from what I\u2019ve seen on Omarchy, I think we can automate a ton of the development and debugging of the system- \u2026 to the point where I can just review that email once a day, and then make the decisions once a day. Yep, goes, no, goes, in, out, and whatever. We\u2019re not there yet, but I think we\u2019re gonna get there.\n \n \n Lex Fridman \n (02:46:36) \n Well, can you sort of, like, empathize with the feeling that, say, a young person sitting in San Francisco, everybody around them is not sleeping. They\u2019re obsessed. They\u2019re programming nonstop. And there\u2019s this kinda sense that AGI is gonna be here any minute. There\u2019s, everybody knows somebody who\u2019s made millions of dollars because they sold a startup. And so there\u2019s a kinda feeling like you shouldn\u2019t be sleeping. You should be constantly managing agents and programming and building, and that feels like a moment in time we\u2019ll look back at.\n \n \n DHH \n (02:47:14) \n It\u2019s the birth of a new paradigm. It\u2019s always messy.\n \n \n Lex Fridman \n (02:47:16) \n It\u2019s always messy.\n \n \n DHH \n (02:47:17) \n It\u2019s always exhausting. There\u2019s no other way around that. And by the way, hasn\u2019t San Francisco always been like this? I remember the dotcom boom years, and everyone was just the same, and then it was all mobile, and before that it was the gold rush. So I think that\u2019s probably- \u2026 just San Francisco. It attracts- \u2026 the kind of individuals-\n \n \n \n \n DHH \n (02:47:34) \n \u2026 who would think, \u201cI don\u2019t have time to sleep.\u201d I\u2019ve not been on that track, generally speaking, but I will admit that the last three months have felt more exhausting than any other project I\u2019ve done in the last five years, since the Hey launch. That was the last time we had a crazy exhausting launch.\n \n \n Lex Fridman \n (02:47:55) \n Do you think you can keep going at this pace?\n \n \n DHH \n (02:47:57) \n No, no, no, no, no. This is not sustainable at all.\n \n \n \n \n DHH \n (02:48:00) \n But I also, I can see the end of the tunnel.\n \n \n DHH \n (02:48:02) \n Because the automation, which by the way, everyone is building. This is so hilarious. Like, everyone is building their little gas town, their little agent coordination, their setup. This is all gonna be solved. We\u2019re not all gonna have to build our own coordination harnesses. Of course we\u2019re not. And in fact, I\u2019m a little surprised that it\u2019s gone this long, that there\u2019s not more of it has been sucked up by the major labs. You\u2019d think that they\u2019d just built this stuff in. But some of it is also this is what a new domain looks like. I remember for a hot moment when JavaScript kind of came to realize its own power, and there was basically a new JavaScript framework every five minutes, right? Like, a lot of churn. This is what happens at the advent of a new paradigm.\n \n \n DHH \n (02:48:41) \n So I think it\u2019s natural, and I think it\u2019s gonna settle down, and I can already see the light at the end of the tunnel. So I don\u2019t mind a sprint. In fact, I welcome it now. I like this notion that most time is calm, but then occasionally you gotta climb a mountain.\n \n \n Lex Fridman \n (02:48:56) \n If we look back to the DHH a year ago, what advice would you give to that guy?\n \n \n DHH \n (02:49:03) \n I wouldn\u2019t wanna spoil it.\n \n \n \n \n DHH \n (02:49:04) \n What a thriller that this has been. I mean, if you would\u2019ve scripted this, I\u2019d be like, \u201cThis is so far-fetched. Get out of here. Unbelievable,\u201d right? I wouldn\u2019t wanna know anything. The fact that this rollercoaster and this acceleration has been absorbed in real time by everyone, no one knew, right? Like, some had premonitions that were a little better than others, but no one knew, not exactly the way- \u2026 it was gonna go and how fast it was gonna take off. I mean, it\u2019s, it\u2019s the show of a lifetime. I mean, the meme cinema- \u2026 absolutely applies-\n \n \n Lex Fridman \n (02:49:41) \n Yeah, it\u2019s incredible.\n \n \n \n \n Lex Fridman \n (02:49:43) \n That said, what do you think DHH from a year from now will be like? I mean, what are the possible versions?\n \n \n DHH \n (02:49:52) \n That all our dreams are coming true. Like, this is the AI maximalist abundance argument. That once Elon\u2019s robots have the level of intelligence and AGI that I\u2019m experiencing on some of this development stuff, the world is gonna be so unrecognizably different that we can\u2019t even imagine it. Now, I don\u2019t spend any time thinking about that because I think that is the way to the AI psychosis. So therefore, I just immerse myself in the moment and get absolutely the maximum out of it, and-\n \n \n \n \n DHH \n (02:50:30) \n Just have fun. I mean, I get it why people are finding this exhausting. Just keeping up. Not even using agents, just, like, being like, \u201cWhat\u2019s the latest model? What tool are we using now? Is it TMox? Is it a Herder? Is it none of it? Is it agents?\u201d Whatever. I get it, but I also again think we should be so blessed. You are alive in this moment where decades are happening- \u2026 in weeks.\n \n \n Lex Fridman \n (02:50:54) \n Quick bathroom break.\n \n \n AI video generation and filmmaking\n \n Lex Fridman \n (02:50:57) \n I\u2019ve been generating a lot of video recently with Higgsfield, and so they became a sponsor. Created a racing video. Wanted to get your opinion on it.\n \n \n DHH \n (02:51:06) \n Oh, yeah? Oh, let me see. Let me see.\n \n \n \n \n DHH \n (02:51:10) \n If they downshift on a straight like they do in the movies, I\u2019ll call it out. That is the worst. Let me see.\n \n \n \n \n \n \n Lex Fridman \n (02:51:19) \n \u2026 this is fully AI generated. So first it goes-\n \n \n DHH \n (02:51:21) \n Wait, what? This is AI? Shit, that\u2019s my suit. That\u2019s my car.\n \n \n Lex Fridman \n (02:51:28) \n Lights go green on a street that never sleeps.\n \n \n \n \n \n \n \n \n Lex Fridman \n (02:51:39) \n But they never came. Is it good? I got a bad case.\n \n \n DHH \n (02:51:44) \n I mean, it\u2019s crazy. You really have to notice in the\u2026 Like, that car didn\u2019t have headlights on and the other one did. And that looked like a 60-year-old version of me. But holy crap, the\u2026 Wow, that was crazy.\n \n \n Lex Fridman \n (02:52:03) \n So there\u2019s parallels here to programming because if you do just straight video generation with no human in the loop- \u2026 you get a lot of weird artifacts- \u2026 and so on, and you can\u2019t really do\u2026 So they do and I highly recommend people go to the Higgsfield YouTube. They have full 90-minute original movies.\n \n \n \n \n Lex Fridman \n (02:52:22) \n And, like, I can\u2019t look away. It\u2019s really cool. \u2018Cause it used to be that it\u2019s just, like, something that just feels like trailers. Like advertisements for something.\n \n \n \n \n Lex Fridman \n (02:52:32) \n This is actually telling stories, like, people\u2019s faces and they\u2019re talking and you\u2019re like, you\u2019re drawn in. It\u2019s not quite where programming is. And so the question is how do you integrate the human into the loop of the filmmaking process? So you have to create the people, and they\u2019re- \u2026 they have to be kept consistent. And then for the images that are kinda like feeding the thing in the creation process, you have to correct through prompting things that don\u2019t feel right. Because when you go from image to video, the stuff that doesn\u2019t feel right will be-\n \n \n DHH \n (02:53:11) \n It\u2019s gonna be zoomed up, right?\n \n \n Lex Fridman \n (02:53:12) \n That\u2019s the stuff that\u2019s- \u2026 going to really create- \u2026 the non-realistic stuff. The question I had for you is, like, do you think outside of programming, can you apply lessons from programming to video creation, to art?\n \n \n DHH \n (02:53:27) \n It\u2019s a good question because I think good art is very dependent on the same thing that good software is depending on: having a vision, having a cohesive idea of what you wanna create. How is it different? How is it novel? How is it going to appeal to people? But I\u2019m not sure that the creative instincts transfer quite as well. I don\u2019t know that I would have the creative insights to come up with good video prompts that would create compelling content. But it\u2019s funny because I have actually, on TikTok, there\u2019s this guy, I\u2019m trying to remember, Goblin something. That creates these\u2026 They are shorts, but not super short, maybe a couple minutes, of really interesting sci-fi vignettes. And I think the guy is actually working on a feature length video right now.\n \n \n DHH \n (02:54:20) \n And I think that this is one of those abundance moments. There are creative people with that vision all over the place, especially in a genre like film, where before it required a $200 million budget to create, I don\u2019t know, a sci-fi movie. Now suddenly, with AI, if you have the right vision for it, you can do that in your bedroom. Now, that story\u2019s been told quite a few times in other domains. It used to be true that if you wanted to record a studio album, well, you needed a studio, and you needed to book that. And now you can record at home on your laptop, and you can do a lot of these things, right? So we\u2019ve had this democratization process happen with plenty of other domains, but film seems to be one of the last ones.\n \n \n Lex Fridman \n (02:55:04) \n And it\u2019s the one that makes, like, me and a lot of people feel weird, right? Like-\n \n \n DHH \n (02:55:11) \n Oh, I\u2019m super excited \u2019cause I think it\u2019s-\n \n \n \n \n DHH \n (02:55:13) \n There\u2019s so much garbage being produced right now. It\u2019s the same argument with the slop, right? So people for a second were worried about slop in programming, and I was telling them, like, \u201cHave you seen the human slop? It\u2019s also pretty bad.\u201d And I think if it comes to studio entertainment, I mean, the bar is pretty low, right? Like, the number of great movies I\u2019ve been dying to see over the past, say, five years, not been a particularly high number.\n \n \n Lex Fridman \n (02:55:42) \n Yeah, the video models are just, they\u2019re able to create incredibly realistic facial expressions.\n \n \n \n \n Lex Fridman \n (02:55:51) \n I mean, I- it just might change the nature of film. I think they\u2019re m- I m- I guess my hope would be that video games and film kind of merge, and that film can become interactive.\n \n \n DHH \n (02:56:06) \n Like the malleable operating system, having the malleable narrative, having malleable entertainment, that man, Game of Thrones, what if we could just forget the last four episodes of slop that was produced-\n \n \n \n \n DHH \n (02:56:21) \n \u2026 and then AI- \u2026 comes up with- \u2026 I don\u2019t know- \u2026 maybe 100 different variations, and someone else watches that, and I get the version that actually finishes off that show in a way that\u2019s fitting.\n \n \n Lex Fridman \n (02:56:34) \n Yeah. Something tells me an individual creator would not create those la- would, would create a different ending for Game of Thrones.\n \n \n DHH \n (02:56:40) \n Oh, for sure. Right? And more just to make the argument that talk about human slop. That was an absolute atrocious ending to perhaps one of the greatest pieces of movie history, I was about to say, but series history, right? It was so good, and it was actually a compelling argument for the fact that you need that strong vision. And as soon as those showrunners had to go without a script, it went off the rails.\n \n \n Lex Fridman \n (02:57:10) \n Do you think\u2026 You\u2019ve mentioned AGI. Do you think we\u2019ve actually achieved AGI?\n \n \n DHH \n (02:57:14) \n No, not in the general definition that it\u2019s in all the things. But have I seen AGI in these glimmers? Absolutely. Especially over the last three months, I\u2019ve seen things where I was thinking to myself, like, \u201cHow would AGI look different from this?\u201d\n \n \n \n \n DHH \n (02:57:34) \n What, what more could there be? And I mean, I can quibble at the margins, but on the big picture, I\u2019ve seen it able to do these long run tasks and coordinate these multiple aspects of it that we\u2019re so far beyond that early moment where I\u2019m telling it, \u201cGo do this, package that.\u201d And now it can carry out a whole task from these vague instruction. That feels very AGI-like, and that\u2019s what\u2019s so inspiring about the moment because you catch these glimmers. And then you go, \u201cWhat if everything was like that?\u201d\n \n \n Lex Fridman \n (02:58:12) \n Doesn\u2019t programming basically unlock the everything? So you can write a pro- like it can generate the program that does the everything.\n \n \n DHH \n (02:58:20) \n So, this is\u2026 I\u2019m in that meta phase right now. So I\u2019m building this Amabot system that\u2019s managing the development of Omarchy, and I\u2019m using AI to build the bot. And I\u2019m seeing that recurrent loop where I started it on, like, \u201cThis is what I want. I want you to be able to have these isolated VM workers and so forth,\u201d and then it can do the self iteration. Then it starts running it. It starts noticing its failures. It starts optimizing it. And a lot of that system, Amabot in particular, AI has driven the majority of the design decisions. Like, I had some vague ideas of where I wanted. I wanted to use this brains and hands pattern where the model is running from where the code is executing-\n \n \n DHH \n (02:59:00) \n \u2026 which is actually a really interesting pattern. Again, Toby alerted me to that, where you have a coordinator that runs the model- \u2026 but it\u2019s manipulating a safe VM where any untrusted code you\u2019re pulling down from pull requests or issues or whatever can\u2019t contaminate the model. And even in that loop, I keep seeing it catch itself, spotting vulnerabilities, where it\u2019s like, \u201cOh, I actually just took some feedback from a test run. If a clever attacker had embedded a malicious payload in the response of the test run, that could have polluted something. I took it for good thing. I better start treating that as outside data.\u201d And you just go like, it is just\u2026 I get why San Francisco is so obsessed with fast takeoff- \u2026 because clearly they\u2019ve seen this 100 times- \u2026 more.\n \n \n DHH \n (02:59:56) \n They\u2019ve seen it with all the models that are in- \u2026 training now, and we keep getting these little glimpses, like the security breach with Hugging Face, where OpenAI was training this model, and the model starts inventing ways, essentially sending smoke signals to itself through embedding messages in a package manager, where you just go like, first, that\u2019s damn clever. Second, okay, that\u2019s also a little scary. And third, how amazing would this be if we could harness this level of intelligence- \u2026 and ingenuity towards productive ends? So this is the it\u2019s so over, we\u2019re so back pendulum that keeps swinging back and forth-\n \n \n DHH \n (03:00:39) \n \u2026 that AI has just delivered that swing back and forth like nothing else. Like barely a, a week goes by where it has to swing one way, and then the next week it swings back the other way. Exhilarating.\n \n \n Lex Fridman \n (03:00:52) \n I\u2019m still at the stage where it makes me truly happy to see those moments of cleverness, and they\u2019re becoming more and more frequent. Just kinda like seeing, oh, like first of all, at the very basic level, when a model gets it. I\u2019ll say a basic prompt, you know, a request, and it doesn\u2019t just do a dumb implementation. It deeply understands, and it\u2026 That was\u2026 That\u2019s a, that\u2019s always a beautiful thing. It\u2019s almost like having a good, like, partner, like program.\n \n \n DHH \n (03:01:22) \n That\u2019s exactly what it is. That\u2019s exactly why it\u2019s so exhilarating and why, as we talked about before we started, it can sometimes be difficult to come out of hyperdrive. That\u2019s also one of Toby\u2019s terms here. That when you\u2019re working directly just one-on-one with agents and you\u2019re not looping in other humans, you go so fast, and when the agents are really understanding your intent, or even better, they revise and improve upon your intent to deliver something that was greater than what you asked for. And you\u2019re doing this with 16 parallel threads. It can be quite difficult to come out of hyperdrive and have to deal with squishy humans. They just, they don\u2019t think as fast. They don\u2019t move as fast.\n \n \n DHH \n (03:02:08) \n And I mean, I\u2019m trying mostly to make light of it, but you can also see a dark version of that where someone does end up in an AI psychosis where they\u2019re not interested in other humans at all anymore. I mean, certainly don\u2019t have that at all, but I do have this sensation at times where I\u2019m really grateful for just the rush of just me and the agents. Like, can I just not talk to anyone else but agents- \u2026 for four hours? Like, that\u2019s an incredible run.\n \n \n Lex Fridman \n (03:02:38) \n Toby is such a fascinating human because he\u2019s running a large company. So I would love to get insights of how AI\u2019s being incorporated effectively in companies. Because, like, for me, humans are pretty slow. And so, and so there\u2019s a temptation to, like, use AI for basically you know, AI running inside Slack, for example, or inside the Basecamp, whatever. And, like, collecting all the information about the human interaction, basically handholding the humans to speed up as much as possible. And like, and then at a certain point you\u2019re like, wait, what is this human\u2026 What are each of us doing that\u2019s not replaceable by AI? And that becomes, like, a scary kind of conversation because a lot of work is kind of replaceable.\n \n \n DHH \n (03:03:37) \n The irony here is it already was. So the number of fake email jobs that currently exist in this world is an absolute epidemic. And-\n \n \n DHH \n (03:03:46) \n \u2026 this, by the way, is not a new thesis. David Graeber wrote this wonderful piece, Bullshit Jobs, back in, I think it was maybe early 2010s. Based on a poll he did in the UK asking people, at that time, would\u2026 I think the question was something like, \u201cWould it make a difference if you didn\u2019t go to work?\u201d For humanity, for mankind. And something like 30-some percent answered, \u201cNo, it wouldn\u2019t.\u201d That a third of workers thought that their job was fake, that it did not produce any worthwhile, valuable outcomes, neither for the community or society or maybe even the economy at all. And this was my argument since day one with startups. This was why we stayed a small company.\n \n \n DHH \n (03:04:39) \n This was why we didn\u2019t want to take VC, because I was so skeptical of the idea that you could get hundreds of programmers, thousands of programmers to do something productive- \u2026 as a collective group.\n \n \n DHH \n (03:04:52) \n And I think what we\u2019re seeing right now with some of the layoffs is a testament to that, not AI. It\u2019s just that during the pandemic, a bunch of over-hiring went on, and now AI is a convenient excuse to slim down. But I do also think AI is going to expose some roles as just not being productive ways for humans to spend their time, and therefore we must come up with new ways of spending their time. And the one, again, I\u2019m quoting Toby for the third time here, but anecdote that he shared me, or observation was, okay, let\u2019s say we all\u2013 we lose half the jobs that are currently happening right now.\n \n \n DHH \n (03:05:34) \n Do you know that Formula 1 employs literally tens of thousands of people just to run cars around in a circle for spectators? Like, that has, that whole circus has no intrinsic value to society besides the spectacle it produces of amusement and entertainment. We just decided it would be really fun for these, what is it, 12 manufacturers to create these fast cars and constantly tweak their little aero improvements, and we will watch it, and we will spend literally billions of dollars doing it and employ tens of thousands of people towards an ultimately deeply frivolous activity.\n \n \n DHH \n (03:06:17) \n So if we are liberated from a bunch of drudgery and fake email jobs, maybe we all get employed as F1 style engineers and drivers and mechanics and masseuses and the other 1,500 jobs that are involved in that, and we come up with new things. I think this is\u2026 Humans are very poor at imagining what exactly the future\u2019s gonna be like.\n \n \n Lex Fridman \n (03:06:40) \n Yeah. Imagine, imagine a future 100-plus years from now, where we look back at this whole span of history, thousands of years where humans worked.\n \n \n DHH \n (03:06:52) \n Funny thing is we\u2019re already doing that, right? Like, the number of fake email job people who sit on a chair all day long and type into a machine who don\u2019t have hard physical labor would be unimaginable- \u2026 to folks in the 1800s, right? So the future is always already here, it\u2019s just not evenly distributed.\n \n \n Lex Fridman \n (03:07:11) \n Unfortunately, as is always the case, the transformation of society required to go from one step to the other is going to have a lot of suffering. And potentially political turmoil and all that kind of stuff. Because you can imagine how many white collar jobs might be lost in this process.\n \n \n DHH \n (03:07:30) \n Yes. And I don\u2019t think you should make light of that. I don\u2019t think it\u2019s funny. But I also do think it\u2019s a necessary component of progress. And I think if we look back upon the Luddites or the folks working the fields, separation has a tendency to reduce empathy. So we have high empathy right now because we know people- \u2026 and we are people who are facing these challenges. But what is our empathy for the Luddites of the early 19th century? Would we wish that all garments were still made by hand weaving? No, we wouldn\u2019t. We want the comfort and accessibility and convenience of being able to buy clothing off the rack. So we must suffer in the moment for future generations to live more prosperous.\n \n \n \n \n DHH \n (03:08:22) \n And that\u2019s always been true.\n \n \n Lex Fridman \n (03:08:23) \n \u2026 I think that\u2019s actually the trick, is to push for progress but have deep compassion for the people who have to suffer. And that\u2019s many of us. To suffer the transformation required to achieve that progress. And sometimes too often in places like Silicon Valley or so on, you can over-focus on, look, the kind of utilitarian view of things- \u2026 and look at the progress. But, you know, considering deeply each individual human that suffers because of that progress is important.\n \n \n DHH \n (03:09:00) \n I do think that suffering becomes more meaningful when you are suffering for a future someone in particular, like your kids.\n \n \n DHH \n (03:09:08) \n And I think this is one of the great tragedies of the moment, is the falling birth rates are making it more difficult for people to be excited about a more prosperous future that is difficult to get to. Because if we\u2019re planting trees of which our children will not sit in the shade, I don\u2019t know, maybe we should just make some lumber and burn it all down, right? I think there\u2019s a nihilism that can sneak into society much more easily when we are not DNA invested in the future, when we are not liable to our own offspring to propel forward. So I think these are bad trends to merge at the same time, that we have these falling birth rates and these problems with coupling, and the fact that we need to keep our chin up about the future and be\u2026\n \n \n DHH \n (03:10:11) \n I mean, you don\u2019t even have to be excited about it in this literal ecstatic sense that I\u2019m portraying here, but I think you should be hopeful, and I think you should be working towards that hope, and I think it\u2019s much easier to do so when you have literal skin in the game.\n \n \n Fatherhood\n \n Lex Fridman \n (03:10:27) \n We\u2019ve been talking about AI agents. Let\u2019s talk about the human side. What do you love most about being a dad?\n \n \n DHH \n (03:10:36) \n That\u2019s a good question. I think the overall love that expands from humans that derive directly- \u2026 from your lineage- \u2026 is very difficult to communicate because I was actually not a particularly big kids person prior to the arrival of my own. And now they are the most interesting people in the world. A lot of the time. Not always. Sometimes they\u2019re also just assholes and so forth. But there\u2019s been such a focus on, \u201cOh, kids are the worst. Look at all the things I can\u2019t do. I can\u2019t go out and drink. I can\u2019t do this. I can\u2019t do that.\u201d Yeah, that\u2019s called sacrifice, and sacrifice is meaningful when you\u2019re doing it for something worthwhile. And what could be more worthwhile than literally the continuation of the human species?\n \n \n DHH \n (03:11:31) \n And that you have a stake in that. Now, there\u2019s also just the sheer joys of watching a human grow from a baby to a toddler to a teenager. My oldest has just become a teenager.\n \n \n DHH \n (03:11:49) \n And I often talk to Jamie, my wife, about this, where the regret I would have of having missed that opportunity on the last day would just be total, unlike anything else. So I do think it is far more satisfying even in the moment that is normally portrayed. Again, as I say, there\u2019s such a focus on all the ways they suck. \u201cOh, you don\u2019t get to sleep,\u201d or whatever. \u201cThey\u2019re annoying. They\u2019re ungrateful.\u201d Yeah, yeah, these things are all true, but the sense of meaning you get from it is unparalleled. And I\u2019m excited about a lot of things. I love building things, and I love hobbies, and I love all sorts of stuff. But creating life with another human that you love is literally the peak experience of being on the planet.\n \n \n Lex Fridman \n (03:12:48) \n What do you think about them coming up in this world full of AI? Is it totally diff- It feels like, maybe I just sound like an old man on a porch, but it feels like a very different world. So I grew up before the internet, but even the internet doesn\u2019t feel like as big of a transformation as this.\n \n \n DHH \n (03:13:10) \n No. But there\u2019s been bigger transformations. I mean, imagine growing or being born in like 1880. Imagine seeing the First World War and the Second World War. Imagine seeing the airplane, the radio- \u2026 television. I mean, the transformation, and I think Peter Thiel makes this argument, that basically nothing has happened in the physical world in quite a long time. We\u2019ve been stagnant. All the development has been in the digital realm. And as important as that is, it\u2019s actually not as consequential as we like to believe it is. Maybe AI will be that thing that is more consequential. It does seem to point that way. But there are other humans at other times who have lived through-\n \n \n DHH \n (03:13:54) \n \u2026 transitions that for them certainly would feel as momentous as what we\u2019re living through. And I think having that sense of history, that you\u2019re not that special in your sense of worries, in your anxieties. This was one of the great revelations of discovering the stoic writings. You have these guys 2,500 years ago dealing with very familiar dilemmas and challenges in their life, and recognizing that we\u2019re not so special in that regard, I think is a great liberation. Because then you think, \u201cOkay, well, we have whatever, 200,000 years of human development before us, and they somehow made it through.\u201d\n \n \n \n \n \n \n Lex Fridman \n (03:14:40) \n \u2026 the thing that makes life worthwhile is basically the same today as it was-\n \n \n \n \n Lex Fridman \n (03:14:46) \n \u2026 2,000 years ago.\n \n \n DHH \n (03:14:47) \n Yes. Love, creation, creativity.\n \n \n Lex Fridman \n (03:14:50) \n Yeah. Same challenges.\n \n \n DHH \n (03:14:51) \n I read this book The Fourth Turning. You check that out. It talks about this notion of history working in cycles, and there are these phases to history, and they name them and so forth. And the specific theory is not as important as just realizing that this is a\u2026 History is a circle, more so than just a straight line. And we are repeating many of the same patterns over and over again. They feel so unique to the moment. And whatever crisis we\u2019re dealing with right now feels so pressing and important until we zoom back a little bit in history and realize that the last one also felt so pressing and important to the people who lived through it.\n \n \n Lex Fridman \n (03:15:40) \n Yeah. You\u2019re making me feel better. So I\u2019m gonna travel across the country and spend a lot of time with no access to the internet, alone with my thoughts. And that\u2019s-\n \n \n DHH \n (03:15:52) \n I think that\u2019s healthy, taking a break.\n \n \n Lex Fridman \n (03:15:54) \n It\u2019s okay. It\u2019s okay.\n \n \n DHH \n (03:15:55) \n I took not a big break, a little break, break from X this summer, three weeks.\n \n \n Lex Fridman \n (03:16:02) \n Yeah, you were\u2026 You disappeared off of-\n \n \n DHH \n (03:16:04) \n I\u2019m surprised you noticed.\n \n \n Lex Fridman \n (03:16:08) \n I went through withdrawals. No. You disappeared for, I don\u2019t know, a few weeks.\n \n \n \n \n Lex Fridman \n (03:16:14) \n Three weeks? What happened? This is after, after Llama.\n \n \n DHH \n (03:16:21) \n I\u2019ve done that before. I just sometimes get the sense that X in particular is the most addictive form of information, entertainment, connection that exists in my life, and anyone that follow my X stream as I was on my flight over here would certainly go like, \u201cThat man is deeply addicted- to the connection.\u201d And yes, guilty as charged, which also means that occasionally I gotta cut it cold, because I do think you can break your brain. And over the summer in particular, for whatever reason, I had somehow managed to steer my feed, my algorithm, towards too much politics, and not enough tech, and not optimism, and not enough building. And even if I agreed with the politics and the angles that were being presented, I just didn\u2019t need- \u2026 that much of it.\n \n \n DHH \n (03:17:23) \n And I think arriving at a place where you just go like, \u201cOkay, that\u2019s enough.\u201d Not that\u2026 Like, it doesn\u2019t have to be something bad. I would also say after eating probably two, maybe three chocolate strawberries, I\u2019d go like, \u201cOkay, that\u2019s enough.\u201d Like, it\u2019s not that I don\u2019t like the chocolate strawberries, it\u2019s just, like, I don\u2019t wanna eat 17 of them. And I think X and the modern algorithms are very good at feeding you freaking 17 chocolate-covered strawberries.\n \n \n Lex Fridman \n (03:17:52) \n I wish I could, uh-\n \n \n \n \n Lex Fridman \n (03:17:54) \n \u2026 control that. Like, \u2019cause there\u2019s\u2026 I would like some politics, 5%. I would like for example, one of the things I don\u2019t like about the tech community is the snark, but I like some snark- \u2026 \u2019cause it\u2019s funny. But I want that to be like 10%.\n \n \n DHH \n (03:18:11) \n Yeah. Yeah. I want it just right.\n \n \n Lex Fridman \n (03:18:13) \n Of my\u2026 Yeah, just right.\n \n \n DHH \n (03:18:14) \n The Goldilocks version of the algorithm. Well, the problem is the algorithm finds out what you respond to- \u2026 not what you say you like. It\u2019s all revealed preferences, and unfortunately, those revealed preferences are not always pretty. I mean, we have this shadow, as Jung talks about, right? All these dark aspects of ourselves, and they are expressed in how long you linger on a tweet, and what you like, and what you don\u2019t like. And the algorithm holds up a mirror- \u2026 and it\u2019s not always pretty, and I just went, \u201cDo you know what? I need a break.\u201d\n \n \n Lex Fridman \n (03:18:45) \n Mm-hmm. I\u2019ve gotten to, I started staying off social media, and I\u2019ve started scraping X. Like, for example, in preparation for this conversation, I\u2019ve scraped all of your tweets.\n \n \n DHH \n (03:19:00) \n That\u2019s actually a method I was using for a while. I was using a tool called, I forget. It was an email tool that would summarize every day, and unfortunately, freaking X cut off their access. Now to get access to the feed, you gotta pay some exorbitant- \u2026 sum that this startup couldn\u2019t pay. So I couldn\u2019t get it that way, and I actually thought, \u201cMan, that\u2019s a miss,\u201d because X has and reveals some of the most interesting thoughts. Some of the most interesting people all the time. But it\u2019s like a damn slot machine, and most of the time you don\u2019t hit the jackpot. You hit, like, something else. And if I could just get it in sort of snack sizes, not like all of it-\n \n \n DHH \n (03:19:44) \n \u2026 it\u2019d be nicer. But that\u2019s not the product. And I think this goes for all of social media. It\u2019s all optimized towards maximal extraction, and if we\u2019re making the parallels here to AI, that is the great fear, right? That once there is a reward function, how much time can I get this human to scroll away? AI and machine learning before we called it AI, has become very, very good at delivering just the stuff that you will find most enraging and most addictive.\n \n \n Lex Fridman \n (03:20:15) \n I think all\u2014I think legitimately engagement-maximizing social media is a huge harm to society. It\u2019s creating this Brave New World situation that I hope is a temporary thing that people just solve. It\u2019s a technology problem.\n \n \n DHH \n (03:20:31) \n So I try to consciously flood the channel in the other direction, and this required a mental shift. I think if you look at my tweets, like, I don\u2019t know, five, 10 years ago, they tilted far more negative. If you look at my tweets now, I\u2019d hope- \u2026 well, I guess we can do a sentiment analysis and see if it\u2019s actually true.\n \n \n Lex Fridman \n (03:20:52) \n No, it\u2019s more good vibes.\n \n \n \n \n \n \n DHH \n (03:20:55) \n \u2026 it\u2019s more positive. It\u2019s more good vibes. It\u2019s more like, \u201cHere\u2019s a bunch of cool stuff I found.\u201d \u201cHere\u2019s what I\u2019m excited about.\u201d \u201cHere\u2019s what I\u2019m building. Here\u2019s some encouragement to other people who are discovering things.\u201d I also accept that that is just not how human brains are built for maximum engagement. Like, there is a, whatever, there\u2019s\u2026 I literally, I think you can quantify that negative sentiments find, what is it, 50% more traction or something? And eventually that just crowds things out. So that\u2019s why so much of this feed is that negative stuff because we respond to it. But you can make a conscious effort to not do that.\n \n \n DHH \n (03:21:33) \n This is one of the reasons I\u2019ve liked TikTok so much. I only use TikTok when I travel, and this actually was the first travel where I didn\u2019t use TikTok, and part of the reason was that even though the internet had high download, it was kind of crap anyway. It wasn\u2019t Starlink, so I couldn\u2019t actually use TikTok on the plane, and I was too engrossed with the agents. But normally when I travel, I watch TikTok. And there it\u2019s such a marshmallow test. A video will scroll down and, like, it\u2019ll activate some sort of primal instincts of rage or lust or whatever, and if you linger on it, you don\u2019t get the marshmallow of science, excitement, art, music, all these other things. You have to be disciplined-\n \n \n \n \n DHH \n (03:22:21) \n \u2026 to instantly scroll away- \u2026 such that the algorithm doesn\u2019t pick up your real preferences. And I find that just to, at the meta game of TikTok- \u2026 to be pretty interesting, and then also just fascinating in themselves. Again, as we talked about, there are a lot of really funny, really talented people in the world that you would never, ever have heard of if it hadn\u2019t been for these platforms.\n \n \n Lex Fridman \n (03:22:46) \n Yep. Who do you think is gonna win the race to AGI in this whole process? We got Google, which is a big surprise to me, to a lot of people.\n \n \n DHH \n (03:23:00) \n I thought they had a good comeback for a second, and then I don\u2019t hear anything about it anymore.\n \n \n Lex Fridman \n (03:23:05) \n Yeah, it\u2019s quite, I mean-\n \n \n DHH \n (03:23:06) \n But I\u2019ve been wrong about this before. I mean, I think in February I had a tweet about like, \u201cOh, all I need is Kimi K2.5.\u201d And in that brief moment, that was true.\n \n \n \n \n DHH \n (03:23:17) \n When I was instructing the agent, I was like, \u201cOh my God, Kimi\u2019s just as good as these other models.\u201d But then the game moved forward and- \u2026 the agents got smarter, and I was no longer satisfied just instructing them, and I haven\u2019t used Kimi a whole lot recently because the frontier models have gotten so good.\n \n \n Lex Fridman \n (03:23:34) \n And somehow Anthropic has been able to, for the most part, stay on top of the programming world.\n \n \n \n \n Lex Fridman \n (03:23:42) \n I mean y- a lot of people argue with Codex being it\u2019s-\n \n \n DHH \n (03:23:46) \n No, I don\u2019t think there\u2019s that much of an argument at the moment. Codex is very good. I use it all the time. It\u2019s my favorite checker. It keeps finding stuff. It\u2019s great. But one of the things I\u2019ve found that I like so much about the current Claude models is they\u2019re really good writers out of the box. GPT is a terrible writer out of the box. Like, the pull requests you get on a blank context that you\u2019re asking it to open, terrible. Like, really awful writing. The pull requests and descriptions and commit messages and so forth I get out of Claude models- \u2026 even though occasionally they can be a little verbose on code comments and you gotta tell them to be succinct. But the writing style is really good.\n \n \n DHH \n (03:24:28) \n To the point that for a while I was just having it reply on pull requests as me, because I was just using my own account. I was like, \u201cThat looks a lot like me, that, like, I could have written that.\u201d But then I went like, \u201cEh, do you know what? That\u2019s disingenuous.\u201d So no, whenever I have an agent post on my behalf, I always have it sign itself like whatever, Claude on behalf of DHH. But they are still ahead. They have the best harness too, in my opinion. Claude Code is still better. I really like OpenCode too, and I really like some features that they have, but overall I was actually sort of weirdly, I don\u2019t know if I was proven wrong, but when Anthropic cut off all the other harnesses from using the subscriptions-\n \n \n DHH \n (03:25:12) \n \u2026 that you can no longer use your Claude subscription in OpenCode. I was kinda pissed because it felt like it was a protective, protectionist move. \u201cOh, we gotta lock you in,\u201d and of course it played into the fact that Claude out of the box still refuses to read agents.md. It refuses to read .agent/skills. It is .Claude skills. It is Claude MD, and it just feels so petty- \u2026 that we now have, we have to put these in directions. So all my projects, they have a Claude MD, and all it includes is a pointer to the agents MD, right? And I just went like, well, if these things are true, if Anthropic is willing to do these petty paper cuts just to be annoying, it\u2019s probably also true that they\u2019re cutting off OpenCode from the subscription just to be dicks about it. And maybe that\u2019s also true.\n \n \n DHH \n (03:26:04) \n But I found that using Claude in Claude Code is actually a good example So I was like, all right, I\u2019m annoyed. I wish it wasn\u2019t so, but this is the other thing I have and social media is so good at this. Some company does something you don\u2019t like, right? Now suddenly you have to swear off everything. Why are you\u2026 Someone just asked me, \u201cWhy are you still using Claude?\u201d I had another example where I really didn\u2019t like what Claude was saying about some writing I had asked it to do before.\n \n \n Lex Fridman \n (03:26:35) \n Oh, yeah, you got there-\n \n \n DHH \n (03:26:37) \n I had an essay it didn\u2019t want to translate into Italian.\n \n \n Lex Fridman \n (03:26:40) \n \u2026 you had a blog post. That\u2019s right.\n \n \n DHH \n (03:26:41) \n The essence of the post that it felt like, well, I\u2019m not gonna translate that because I don\u2019t agree with what you\u2019ve written. Which to me was straight out of Space Odyssey, right? HAL telling me, \u201cI\u2019m sorry, Dave. I can\u2019t do that, Dave.\u201d Open the pod bay. Translate my essay to Italian. And it wouldn\u2019t do it. And I just thought like, well, that\u2019s kind of terrifying. But then I also went, well, why am I asking Claude to translate my essay into Italian? I never asked Claude for that because Claude just does my code. So I could also just accept that maybe Anthropic is run by a bunch of people I don\u2019t share much politics with.\n \n \n DHH \n (03:27:20) \n And they don\u2019t want their agents to be used for that, and I think that kinda sucks, but that doesn\u2019t mean it, I have to stop using it for generating code. And I\u2019ve thought, and I\u2019ve tried to do that with people too, become better at tolerating the fact that not only will I not agree with everything someone has to say, I don\u2019t\u2014 I can even be mad about certain positions or things they say and still go like, \u201cWell, you\u2019re still an interesting person, so I\u2019m still gonna follow you.\u201d And I\u2019ve actually kinda gone back on a number of people where I\u2019m like, oh man, that person really pissed me off because they took this point or another point, and went like, do you know what? That\u2019s stupid. I should be able to accept the full spectrum of someone. I don\u2019t have to like all the shades of it.\n \n \n DHH \n (03:28:06) \n I could just go like, okay, well, on that topic we either disagree or, even worse than that, I think you\u2019re a fucking idiot. And on another topic, you\u2019re a smart and insightful person. Because first of all, of course, it\u2019s a little self-serving. I know for a fact that this is how lots of people feel about me. Like, oh, if he could just stick to talking about these topics, it would be so much easier on my feed because otherwise he\u2019ll say these things I disagree with, and that annoys me greatly. I\u2019m like, do you know what? Human beings are jagged like that, and we\u2019re not fully compatible on all the levels. Nor should we be.\n \n \n DHH \n (03:28:40) \n The world would be so fucking boring if we constantly sought 100% compatibility in our opinions and in our politics. And when it comes to my code generator, which is what Claude is, I don\u2019t need it to share all my politics.\n \n \n Lex Fridman \n (03:28:56) \n And therefore it\u2019s nice to have competition.\n \n \n DHH \n (03:28:58) \n And it\u2019s also nice to have open weight models. So this is, again, ironic that we\u2019re getting this out of China. In fact, I did this test right after the issue with the essay that couldn\u2019t get translated into Italian. I asked Kimi K27, to be fair, through OpenCode, so not the Kimi harness, and inference through Fireworks, so an American inference. I asked it about Tiananmen Square. Like, that\u2019s usually the test, right? Like, all the Chinese models, they don\u2019t wanna talk about that moment in 1989. That was actually how I, that was how I posted. I was like, \u201cWhat happened in China in 1989?\u201d That was the prompt. And K25 7 on Fireworks just answered, like, extremely bluntly- \u2026 With a description and depiction of what had transpired-\n \n \n \n \n DHH \n (03:29:49) \n \u2026 to a point that, like, there\u2019s no way that that would\u2019ve been kosher with the Chinese censors, right? And I just thought, like, isn\u2019t that ironic? I\u2019m running a Chinese open weight model that I can tell or I can ask about 1989 and Tiananmen Square, and in America, I can\u2019t get a frontier model to translate a marginally controversial essay about immigration into Italian.\n \n \n Lex Fridman \n (03:30:16) \n Mm-hmm. I don\u2019t know, you and I might disagree about this actually. I was really deeply troubled that the US government banned a model or pressured Claude and then-\n \n \n DHH \n (03:30:31) \n It\u2019s still very muddy what exactly transpired. Was this sort of petty retaliation for whatever the Department of War? Was this simply Anthropic themselves going like, \u201cOh my God, we discovered a cyber weapon so powerful that we can\u2019t let anyone have it\u201d? Well, okay you can\u2019t be surprised then if the government wants a say on a cyber weapon so powerful it can-\n \n \n \n \n \n \n Lex Fridman \n (03:30:56) \n \u2026 The problem I have with that whole thing is, no matter what happened, it sets a precedent. And so government, when you give it that power, it\u2019s gonna start abusing it. I could see for political means pressuring a model. You know, if it\u2019s a Republican president pressuring models to be, to censor more,\n \n \n \n \n Lex Fridman \n (03:31:18) \n \u2026 Democratic views and-\n \n \n \n \n Lex Fridman \n (03:31:20) \n \u2026 For vice, and vice versa.\n \n \n DHH \n (03:31:21) \n That\u2019s why we should insist on the fact that this stuff, like, I\u2019m sorry, Dave, I can\u2019t translate your essay into Italian. That\u2019s way beyond the line. Like, whatever the sort of root of it is, you have to be a tool first. Now, if the tool is, \u201cHey, can you tell me how to make anthrax?\u201d Okay, fine. Right? But within the jurisdiction you\u2019re in, and certainly in America, free speech is quite absolute. Very, very few limits on free speech, and that\u2019s one of the reasons why the country\u2019s so great. And yeah, I don\u2019t like that. But on the other hand, I\u2019m also a big fan of market competition. So while I don\u2019t like this about Anthropic models, I appreciate that I could go next door to Groq and get my task completed.\n \n \n Lex Fridman \n (03:32:11) \n Do you think outside of blog posts, do you think there\u2019s some legitimacy, like can you steelman Dario\u2019s case that we should be careful about the release of these models?\n \n \n DHH \n (03:32:25) \n Is AI safety important? Yes. Are\u2013 do these models have capabilities that could be seriously harmful in the production of biological weapons or otherwise? Yes. Is it fair to have some ground rules on that? Yes. But then don\u2019t squander it by denying the translation of an essay, because then you erase the whole thing and you- \u2026 bias everyone towards thinking like every guardrail you put up is gonna be bullshit.\n \n \n Lex Fridman \n (03:32:55) \n Yeah. The security piece of it is tricky because it\u2019s so good at finding vulnerabilities.\n \n \n DHH \n (03:33:01) \n Shockingly\u2013 I was about to say annoyingly, but that\u2019s not actually true because what it\u2019s finding, at least when you\u2019re using in defensive ways, is it\u2019s finding real issues that a clever hacker could exploit. But nothing has caused as much stress, to be fair, at the technical team at 37signals, like the fact that these latest batches of models are exceptionally good at finding these issues. So suddenly there\u2019s just this seemingly endless parade of security vulnerabilities that need to be patched and fixed and sorted and so on. And the end result is we end up with vastly more secure systems. But the road there is pretty rocky.\n \n \n Lex Fridman \n (03:33:41) \n Do you think it\u2019s\u2013 \u2019cause you said seemingly endless, do you think it\u2019s a finite set, like we can get to the bottom of it? So just make the systems much, much more secure, where we get back to this semi-stable state where mostly it\u2019s very difficult to find security vulnerabilities.\n \n \n DHH \n (03:33:59) \n I do think that all the aggressive tendencies of these models, or not even tendencies, uses of these models are also the defensive ones. So if you\u2019re good at finding an exploit for exploits, you\u2019re also good at finding exploits for defense. So in that sense, it does feel like we\u2019re spy versus spy here, black and white, but with the same capabilities and with the same actions. But while we still have to steer these agents somewhat and verify and vet the work, it\u2019s creating a lot of intense moments for security teams around the world. And of course, the more concerning part is all the teams that are not stressed, like they just don\u2019t know, right?\n \n \n DHH \n (03:34:45) \n Like if you\u2019re a team right now and you\u2019re not dealing with a bunch of patches, it\u2019s just because you\u2019re blind, and it means that your adversaries quite probably have access to techniques that can get into your system.\n \n \n Lex Fridman \n (03:34:59) \n And we should also say one of the big security vulnerabilities in the world is the human side. You know, social engineering, and that\u2019s extremely concerning with increasingly intelligent AI systems, you know, being able to write emails and phone calls and faking voices, thinking about folks less tech-savvy.\n \n \n DHH \n (03:35:19) \n This stuff is gonna come, and it\u2019s gonna come in bulk. But again, I choose to be optimistic about it that-\n \n \n \n \n DHH \n (03:35:25) \n \u2026 whatever these models can do aggressively, they can also do it defensively.\n \n \n Lex Fridman \n (03:35:29) \n So like you said, recently released Amate 4, Quattro. And you said it\u2019s one of the greatest software releases in my professional career, which is a hell of a statement.\n \n \n DHH \n (03:35:42) \n It\u2019s a hell of a milepost for what\u2019s possible at the moment. And the reason I said it like that is not that I haven\u2019t worked on anything else of note. I\u2019ve worked on a lot of things of note, a lot of things that I\u2019m very proud of. There are very few other projects that I have worked on that in the amount of time that I\u2019ve spent on Quattro, which is really three months, I have been able to accomplish as much, fulfill as many hopes, desires, wishes. It truly is like the metaphor we used earlier about finding the genie in the bottle-\n \n \n DHH \n (03:36:24) \n \u2026 and having an unlimited number of wishes, more or less, for getting everything I wanted out of a system. So it\u2019s strange because it\u2019s also perhaps the major software release that I\u2019ve worked on the least in terms of lines of code. I have the least number of personally hand-chiseled lines of code in Quattro than I\u2019ve had in any major release I\u2019ve ever done. And yet, I am insanely proud of what we\u2019ve accomplished. I would not have guessed that that was how it was gonna play out, that you could divorce the sense of achievement and make it be like a team victory.\n \n \n DHH \n (03:37:08) \n Where other releases, I was very happy for my personal contribution because, like, I was the one chiseling out the lines of code, and then I would push it out there, and then later on, of course, these projects, Rails in particular, would grow into be huge open source projects with thousands and thousands of contributors, and then it certainly was a team effort. But in the beginning, all code bases that I\u2019ve\u2013 almost all of them that I\u2019ve worked on, I started, like, just myself. And Quattro started as a team effort right from the get-go, and where I was just the coach. I was telling the agents where to go, and we would go. So to be able to find satisfaction in that actually gives me great hope for retirement in some way, that there is a phase after playing that you can enjoy.\n \n \n DHH \n (03:37:56) \n And I know that\u2019s not easy. You look at sports and the number of professional athletes who really struggle to cope with the fact that their career is over and have to find a new identity as someone who doesn\u2019t play football or race Formula 1 cars. I mean, it\u2019s more the rule than the exception. So I wasn\u2019t so sure about myself. I wasn\u2019t so sure about having to give up or not even having, wanting to give it up. Wanting to play another level of the game and finding not just equal satisfaction, but more satisfaction in it.\n \n \n Linux will win the desktop\n \n Lex Fridman \n (03:38:34) \n Do you\u2026 You\u2019ve been pretty optimistic about Linux. Do you think legit Linux can increase its adoption? \u2018Cause, you know, for many years the meme is, you know-\n \n \n DHH \n (03:38:45) \n Linux on the desktop- \u2026 every year. Next year again.\n \n \n Lex Fridman \n (03:38:48) \n Yeah. But it seems like, at least the case you\u2019re making, the energy you\u2019re putting into it, you could see a vision where it takes over because agents love Linux.\n \n \n DHH \n (03:38:58) \n Not only can I see it, I find it to be the most probable outcome at this point. It is simply too well-suited for the moment. And again, as we talked about, it\u2019s a great irony that all the flaws of Linux, the arcane config files, all the strange error messages and so on should just so happen to be the perfect thing for an agentic operating system. But so it is, and we should rejoice that fate has this irony, and just laugh at it. How\u2026 It\u2019s so funny to me that all the advantages that Apple had for the longest time- \u2026 that they produced this really locked down curated experience should become their greatest drawback now. If you\u2019re into agents and development and working with all this stuff, the Mac is just a hostile place to be. It just has walls all over the place.\n \n \n DHH \n (03:39:59) \n And Linux is simply open through and through. And it\u2019s also interesting because that was not obvious either. If you look at the way a lot of the Linux communities, open source communities, have reacted to AI, it is not universal love. I would argue that the majority of them are actually, if not skeptical, then outright hostile- \u2026 to AI. Now, the saving grace is that the BDFL himself, Linus- \u2026 Torvalds, just wrote a few weeks ago that he actually welcomes AI.\n \n \n \n \n DHH \n (03:40:35) \n He wants to steer it. He wants to make sure it\u2019s good and whatever, but I- the line was something, if you think Linux is an anti-AI project, think again, and you should just do the open source thing and fork it, because we\u2019re gonna use AI. And you see these graphs of the number of AI contributions going into the kernel, and it\u2019s a parabolic curve. So Linux is leaning hard into this. And it actually in, in other ways it was surprising that it didn\u2019t. Linux conquered everything else. All the AI infrastructure that everyone runs off, it\u2019s all running on Linux.\n \n \n DHH \n (03:41:11) \n All the systems, all the servers, everything is Linux. So it was kind of a curiosity that the desktop and the personal computers we were using really hadn\u2019t been captured, but clearly it was just waiting for this moment. Linux spent the time from \u201991 to now waiting for agents to fully flourish as an end user operating system.\n \n \n Lex Fridman \n (03:41:35) \n I hope it takes over. I mean, it\u2019s perfect. It\u2019s perfect for agents, but it\u2019s hard for people to switch.\n \n \n DHH \n (03:41:41) \n I think it\u2019s hard for people to switch when there\u2019s not a compelling reason to do so. This was one of the driving design goals for Omarchy. It was not gonna be Temu Windows or Temu Mac. It was not just gonna be a cheap copy where we try to make it as familiar as possible and then kinda worse. I mean, if I was being unkind, I would say that\u2019s what Ubuntu has tried to do. And there\u2019s something in that, or there was something in that at least, where if you make something really familiar, you can get some people who don\u2019t wanna learn something new, and then maybe they\u2019ll be compelled about your, whatever, free as in speech, not as in beer stuff. And it just didn\u2019t pan out that way. People just didn\u2019t care. They just wanted whatever was better.\n \n \n DHH \n (03:42:27) \n And I think this is why Linux now has the opportunity to win. Because as an agentic operating system, as a malleable operating system, you can tailor to your desires, it is unparalleled. And that is so compelling, and I can see it now. Just since the release of Quattro, the amount of people who have embraced that aspect of it, the malleability, started making their own things, sharing screenshot of it, and being absolute just flooded with positive emotions and dopamine- \u2026 is, that\u2019s the product market fit signal, right? Because what a lot of these individuals are experiencing, they\u2019re experiencing the very high of what the best of being a programmer is like.\n \n \n DHH \n (03:43:16) \n You tell the computer these arcane commands, this rigid logical structure, and it produces what you want. Now suddenly, you sit down and in plain English ramble for 20 minutes. \u201cOh, it\u2019d be great if it did that. Actually, no, that\u2019s not right. Let us\u2026\u201d And out comes software. Out comes an operating system shaped in your image. I think that is one of those experiences that if you have it\u2026 If you feel that power in your hands, it\u2019s very difficult to go back.\n \n \n Lex Fridman \n (03:43:53) \n And I think if you get it right, it\u2019s one of those things that will seem obvious in retrospect.\n \n \n DHH \n (03:43:58) \n Yes. This is how computers were always meant to be. This is how computers started. The Commodore 64, you turn it on. First of all, it boots in a second. Amazing. Second of all, it boots straight into BASIC. It was the malleable computer from the get-go. It just so happened that you needed to know hieroglyphs- \u2026 to be able to take advantage of that. Now we\u2019ve translated it. Agents have given us the Rosetta Stone, and you can simply just speak your desires and- And so they are\n \n \n Lex Fridman \n (03:44:28) \n Beautifully put. Since you mentioned Linus Torvalds, and since you\u2019re working on a Linux distro, you get to have contact with the kernel. What can you say about the genius of this one guy that helped create and grow this ecosystem?\n \n \n DHH \n (03:44:45) \n The perseverance and the commitment and the longevity is truly remarkable. Again, Linus started in \u201991. He hasn\u2019t stopped. I don\u2019t know, he\u2019s probably taken some breaks along the way. It doesn\u2019t seem obvious that there was that many of them. It seems like he just really likes the flow of doing kernel development. He really likes steering where this 40 million line code base is going. He really likes aggregating the net capacity of hundreds of thousands of contributors, and we, as beneficiaries of that, must just sit back and marvel. I mean, at this point, Linux is a system. Like, if Linus stepped out for a while, it would continue to go, but it\u2019s still one of those protect this man at all cost.\n \n \n Lex Fridman \n (03:45:36) \n And he\u2019s still open-minded enough to be able to evolve and- \u2026 accept AI into this.\n \n \n DHH \n (03:45:41) \n Yes, not only AI, they\u2019re letting Rust into the kernel too. I know that was a controversial one too. He has this capacity to look at new information- \u2026 and arrive at different conclusions.\n \n \n Lex Fridman \n (03:45:54) \n What do you think about his style of communication that is akin in some ways to your own, where he can be sometimes a little spicy?\n \n \n DHH \n (03:46:02) \n I think the world has gotten too bland. We need a little spice.\n \n \n Lex Fridman \n (03:46:07) \n Amen, amen to that.\n \n \n DHH \n (03:46:08) \n Now, I think there\u2019s a fine line between being spicy and mean. But it has to be in a good spirit. You can be harsh, let\u2019s put it that way. If you\u2019re trying to impart a lesson to someone you believe could actually learn it- \u2026 I actually don\u2019t think it\u2019s always the worst thing in the world to be a little harsh. Sometimes that lesson sticks, and I think if you talk to people who\u2019ve worked for difficult leaders, Elon, Steve Jobs, maybe even Bill Gates, many of them afterwards would go like, \u201cThese were some very difficult years- \u2026 but also some of the very best.\u201d \u201cThey got the best out of me.\u201d Now, I don\u2019t think it\u2019s my style. I mean, maybe people I interact with feel differently, but in personal interactions, I try and get to the outcome and the lesson in a less abrasive way.\n \n \n DHH \n (03:47:06) \n But I also think we need some spice in this world. We need unreasonable people, and Linus has justification to be unreasonable. In many instances, the weight of the world is literally on that man\u2019s shoulders. The Linux kernel runs the entire civilized society. If the Linux kernel suddenly disappeared tomorrow, nothing would work. So if he\u2019s not harsh when that\u2019s at stake or at risk, like when ever would you be harsh? So I feel sometimes also there\u2019s a proportionality to the size of your mission. This is also why I think these stories about Elon sometimes perhaps being a little bit difficult are easier to rationalize. It\u2019s not even that you condone. You can separate those two things and go like, \u201cDo you know what? It\u2019d probably be\u2026\n \n \n DHH \n (03:48:01) \n Maybe he would have an even easier time getting to his objectives if Linus or Elon or anyone else in that position was a little nicer.\u201d But also who am I to judge that? Like, when you have the weight of that on your shoulders, either firing little rockets at Mars and transitioning all of us to electric cars and whatever, or you\u2019re responsible for the damn Linux kernel that runs billions of devices. Do you know what? It is not my place to call a sort of a code of conduct on you. Again, is there a line? There\u2019s a line for everything. There\u2019s\u2026 Something can\u2019t\u2026 But, like, a few harsh words on a mailing list? No. It\u2019s actually a beacon in some regards that, like, there are standards-\n \n \n DHH \n (03:48:46) \n \u2026 and if you fall below those standards when the stakes are as high as the Linux kernel, ah, you might get ridiculed in public. That\u2019s a liability you should be willing to endure if you participate in this project.\n \n \n Lex Fridman \n (03:48:58) \n Yeah, for high-impact things like that, it\u2019s probably good to prioritize the pursuit of excellence versus the pursuit of niceness.\n \n \n DHH \n (03:49:07) \n And even more to the fact you\u2019re just not gonna get those kinds of individuals being fully rounded- \u2026 plush, always the perfectly polite people. I mean, Elon has this line, \u201cDid you also think I was gonna be a nice\u2026\u201d Or, or what is it? The-\n \n \n Lex Fridman \n (03:49:23) \n A normal dude or some-\n \n \n DHH \n (03:49:24) \n \u2026 normal chill dude. What? No, of course you\u2019re not gonna be a normal chill dude. What would the world gain if we got one more normal chill dude and we had to trade Elon? Holy fuck.\n \n \n Lex Fridman \n (03:49:37) \n I also like diversity of people. I like nice people. I like assholes. I like different ideologies being well-represented, and that\u2019s why freedom of speech works. They clash, and we figure stuff out.\n \n \n PewDiePie\n \n Lex Fridman \n (03:49:51) \n There\u2019s another guy. What do you think about PewDiePie using Linux? He\u2019s an Arch person, no?\n \n \n DHH \n (03:49:56) \n He\u2019s an Arch person. He\u2019s got the most gorgeous arc. So the world\u2019s biggest streamer for a while, I remember my kids watching some of his Minecraft videos, and then he just gets fed up with that, does the wholesome thing. Marries-\n \n \n Lex Fridman \n (03:50:14) \n Yeah, family man now\n \n \n DHH \n (03:50:15) \n \u2026 a family man, kid, moves to Japan, and then gets so hardcore into- \u2026 first Linux, then Arch, then Ricing. Did you see his, uh- \u2026 rice- \u2026 that was Chernobyl rice? Absolutely best tier incredible stuff. And then he also becomes a goddamn AI man of the moment building all of these systems, and you\u2019re just going like, that\u2026 What an inspiring story of evolution. Like, you can go from being funny guy on Minecraft streams to, \u201cOkay, this is my life now. I\u2019m building AI clusters and doing the Council of AIs,\u201d and we all get to spectate? I mean, what a treasure.\n \n \n Lex Fridman \n (03:50:58) \n So I guess, I mean, he\u2019s a pretty good embodiment of what the future of software, or what the future of building looks like, right? \u2018Cause he\u2019s a non-programmer technically- \u2026 becoming a programmer.\n \n \n DHH \n (03:51:10) \n Hugely inspiring, and should be hugely inspiring to others who sit in that moment like, \u201cWell, I don\u2019t know quite how to program. I don\u2019t know this and the other thing.\u201d Okay, but if PewDiePie can build the Council of AIs with his bespoke hardware here, then maybe you can also start on things. And I think this is actually an important general point, that we need role models. We need people to inspire others, to push further, and reveal that the perceived boundaries, they\u2019re not as fixed as you think they are. And every one of those people, they started out in a situation not too dissimilar from yours. Again, that doesn\u2019t mean everyone is gonna be PewDiePie or Elon or Linus. There is not an equal distribution of talent or even intelligence.\n \n \n DHH \n (03:52:02) \n And we can come to terms with that and still be inspired by these people. Like in the physical realm, I feel like we don\u2019t have any problems with that usually. Like, oh, someone\u2019s really good at football, they\u2019re really good at basketball, they\u2019re really good at something, and we go like, \u201cOh my, that\u2019s amazing. I mean, I wish I was, but I\u2019m never going to be.\u201d Sometimes I feel like we struggle a little more when it comes to the intellectual realm.\n \n \n Lex Fridman \n (03:52:23) \n Well, for a lot of people, you\u2019re an inspiration. The way you\u2019re doing with Omarchy\u2019s\u2026 It just shows that you can dream big and really build. It\u2019s great.\n \n \n DHH \n (03:52:32) \n I do like that part of it, and then I accept the part that there\u2019s also a million people who think I\u2019m the biggest idiot on the earth and I don\u2019t actually know anything, and Omarchy\u2019s just a bunch of dot files, and-\n \n \n Lex Fridman \n (03:52:42) \n Do you have some critics on the Omarchy side?\n \n \n \n \n \n \n DHH \n (03:52:46) \n And a lot of it, I mean, is as often happens when someone comes in from the outside. I mean, I\u2019ve been using Linux for two and a half years. That\u2019s not very long. Plenty of people in that community who\u2019ve literally been running Linux since it was hard and difficult, and you had to walk uphill both directions against the wind in the snow barefoot. And I think there is a certain kind of nerd- \u2026 who\u2019s threatened when the community expands, and it suddenly allows more in. And do you know what? I actually\u2026 I can\u2019t begrudge that fully. I think it\u2019s fair for certain nerds to feel like, \u201cThat\u2019s my space.\u201d\n \n \n DHH \n (03:53:23) \n Like, now you\u2019re changing it into another space. I don\u2019t wanna do that. Like, I don\u2019t wanna change the Arch space at all. Like, I don\u2019t really think of Omarchy as being directly part of the same community. Because the people who were attracted to Arch in the first place were the ones who were attracted to building everything by hand themselves.\n \n \n DHH \n (03:53:41) \n Doing it the hard way because they like to do it hard, and I\u2019m building the polar opposite of that. So I just wish that those people could go like, \u201cOkay, well, this is not for me. I wanna build it the hard way. I wanna put in 500 hours to ricing my own Arch distribution,\u201d and I think you should. I mean, actually, I said\u2026 I mean, I did it. This is what Omarchy is, me pouring in literally, at this point, I don\u2019t know, 3,000 hours into this distro. You should do that if you\u2019re so inclined. I think it\u2019s wonderful. But also do realize that we can share the same technical underpinnings and then operate two very separate-\n \n \n DHH \n (03:54:19) \n \u2026 communities with very different aims, goals, aesthetics, morals even, and coexist, and it can be beautiful. And I mean, that\u2019s not always what I get. Sometimes I get the other thing. But at this point I\u2019ve been in the game long enough to not just be at peace with it, but actually be welcoming of it. I take it as part of a barometer that things are working. When I\u2019m working on something that works, I usually get some people like it, and then I get this balance of the universe that we talked about last time, where an equal but opposing force must be present on the other side of the scale to hate upon it. And therefore, I at this point just smile a little. Like, it\u2019s okay.\n \n \n Lex Fridman \n (03:55:02) \n I mean, that said, this, it\u2019s quite a popular distribution Omarchy, at this point.\n \n \n DHH \n (03:55:08) \n And do you know what\u2019s interesting about that is, like, this is the second attempt. So my first attempt was Omakub. And I was really happy with that, and I put it out there, and there it found a community of maybe a few thousand. Like, and that was that, and then it just sort of petered out because the level of ambition couldn\u2019t attract more. And then the second time, because I didn\u2019t fucking stop, right?\n \n \n DHH \n (03:55:30) \n Like, I just kept going. I just kept building. Then I put out Omarchy, the first version. Again, quite niche because it was this\u2026 It was like Git checkout. It was a little bit difficult. You had to set up Arch first. I was\u2026 Well, I would talk you through how to set up Arch, and then you could do the Omarchy thing, right? And then we got an ISO that meant you could just install the whole thing by yourself. And then we just fixed all the problems. This is the other thing. If you keep going, eventually the thing will just work. This was the charge against Linux itself for the longest time. Well, you install Linux and then this doesn\u2019t work, and my speakers don\u2019t work, and my trackpad doesn\u2019t work. Yeah, okay, but just leave Linus at it for like 30 years. He\u2019ll get to it.\n \n \n DHH \n (03:56:10) \n He\u2019ll get to all of it. And at this point I find it hilarious that on, I would probably argue the majority of computers, it\u2019s wiped. You install Linux, all of it works.\n \n \n DHH \n (03:56:20) \n You install Windows, good luck hunting down the drivers you need to get that piece of hardware working. Because they have very different philosophies. The reason Linux is 40 million lines of code is Linus just puts all the drivers in the kernel. So it has everything out of the box, and Windows doesn\u2019t. A different, different approach to it, right? So sometimes the things that look like a toy, things that look broken at first glance, they are, right? They\u2019re for people who like having fun. We call those early adopters. They play with the toys, and then the toys evolve, and then they get better, and then they get better. And at some point, they\u2019re so damn good that early adopters feel like they\u2019re invited to the party.\n \n \n DHH \n (03:57:03) \n And if the early adopters keep chipping away at it too, eventually you get to the late majority, and then you win. And we are well on the way with Omarchy towards that destination. Like the growth that the distro has seen, not just since Quattro, but since version three, it\u2019s just like, you know, the standard hockey stick curve. You go like, there\u2019s nothing, there\u2019s nothing, there\u2019s nothing. You\u2019re just toiling away. You\u2019re just improving, you\u2019re improving, improving. And then you hit this magic inflection point, and it\u2019s impossible to predict when that\u2019s gonna happen. Same thing with startups. When is it gonna happen? When are you gonna find that product market fit? Then suddenly it\u2019s there, and then it goes, just goes and off the rocket ship you go.\n \n \n Lex Fridman \n (03:57:42) \n Well, this kind of vision of making it agent first, of making it-\n \n \n DHH \n (03:57:45) \n I do think that\u2019s the\u2026 We were already on the up, and then it changed to vertical. Because the other thing, and this was what I learned with Omakub and Omarchy first too. Omakub was stylish, nice, but familiar, is we\u2019re still using a conventional desktop metaphor. You were dragging your windows around. And I thought at the time, as I think lots of people do in the Linux community, like this is what we need. And then I just go like, okay, well, for whatever re- let me do the nerdy thing for a while. I\u2019ll build Omarchy, and maybe it\u2019ll just be for me, because I mean, it seems like there\u2019s about five people who like tiling window managers. That\u2019s not true. But as a percentage of total computer users, it\u2019s a tiny, tiny minuscule part. And then I go like, well, this is amazing.\n \n \n DHH \n (03:58:29) \n Totally different, but amazing. I put that out. Instantly, Omarchy had far greater traction than Omakub ever did. Because it was not the same. Because it was totally different, and it presented a different vision for what a computer could be and how it could feel like, and that attracts people. We are now getting sort of the exponential of that. As I said, a lot of people in the Linux community are at best skeptical, at worst hostile to AI and to agents. Omarchy is the first distribution, at least that I\u2019ve seen on, on a scale that matters, that just goes like, \u201cNope, we\u2019re really into it.\u201d\n \n \n DHH \n (03:59:04) \n I mean, you can still use Omarchy. You don\u2019t have to use an agent. If you dismiss that first notification to set up a default agent, you\u2019re not gonna get any of the stuff we talked about. You\u2019re not gonna get the crash watcher. You\u2019re not gonna get any of this stuff. You\u2019re not gonna get\u2026 You can hand chisel all your plugins. Totally possible, right?\n \n \n Future of programming\n \n DHH \n (03:59:19) \n But the reason people are excited is because I don\u2019t have any reservation about saying, \u201cThis is where we\u2019re going. I think the future of the personal computer is the malleable computer, is the agentic computer. I\u2019m gonna go all in on that.\u201d And then anyone who\u2019s excited about a similar trajectory for computers, come along. Plenty of, plenty of room in the car. In fact, one of the things I pride myself on is not holding a grudge to people who come around. I had a bunch of people just last week going like, \u201cYeah, I used to think Omarchy was kind of shit. That was just this crappy-\u201c\n \n \n DHH \n (03:59:56) \n \u201c\u2026 dot files collection. And now I tried Quattro and awesome. I like it now.\u201d That to me, I live for the long argument. I live for this argument where you plant the seed like two years ago, and then two years later they go, \u201cGoddammit, this son of a bitch was right.\u201d\n \n \n Lex Fridman \n (04:00:14) \n Yeah. You do realize if you continue to be as successful as you are, OpenAI and Anthropic are gonna roll in and try to do their operating system or to buy, like with Peter with OpenClaw, there\u2019d be a gigantic check.\n \n \n DHH \n (04:00:31) \n Yeah. I think the good thing about my current position is I really don\u2019t need the money. So I get to build these things purely for my own enjoyment and purity of vision, which has this weird quality where sometimes when you stop caring about what everyone else thinks or wants or whatever, and you just pursue this singular vision, it ends up becoming far more appealing. That it\u2019s when you try to do, oh, I should do this because these people want that, or I should do this and this and so on, you start watering the thing down and suddenly it\u2019s a bland ball of nothing.\n \n \n DHH \n (04:01:08) \n So I mean, again, the thing too here is directionally, I just want computers to be this way. So if Omarchy should end up being just a little footnote in history that it was part of this early movement of the agentic OS and the malleable computer, that\u2019s also okay. I\u2019m completely at peace with that. As long as I get to have a computer that is as fun to work with as Omarchy- \u2026 that\u2019s great. I\u2019ve had the same thing with Ruby and with Rails. From very early I went, do you know what? I love Ruby. It\u2019s such a great programming language. But if a better programming language comes along, I, me, I\u2019m gonna use it.\n \n \n DHH \n (04:01:45) \n If a better Rails comes along, I\u2019m gonna use it. And I think, and I love Ruby, and I will continue to write Ruby even just for the sheer fun of it, like I would go ride a horse. I have in the stables, even though I have a Model Y in the garage. The language now is English. It\u2019s a cliche, but it\u2019s also true. Like I\u2019ve been programming in English for the last three months. I\u2019ve been reading a lot of code, but I\u2019m programming in English. I\u2019m telling the computer what to do, and I\u2019m using natural language, and it is shockingly even more delightful. If there is one programming language more beautiful than Ruby, it is the English language.\n \n \n Lex Fridman \n (04:02:28) \n And now we get to carry it into the natural evolution of human civilization, which is our cyborg future.\n \n \n DHH \n (04:02:36) \n Beautiful. It\u2019s beautiful. And I mean, the reason I say this in part is I always loved writing. I always just loved the English language- \u2026 for the sheer beauty of it, for the sheer intricacy, for the depth of it. I mean, Ruby is a very expressive programming language, but it can\u2019t hold a candle to English. I mean, all the poetry and literature in the world that\u2019s been expressed through the English language. I mean, I like a beautiful code poem, but I mean, the real deal in English is just on a different level.\n \n \n Lex Fridman \n (04:03:10) \n And I actually find this is a nuanced point to express, and I\u2019ll probably fail expressing it. But when you prompt a system and you\u2019re over-specific, it will listen to you too carefully- \u2026 and follow the instructions. And I\u2013 But nevertheless, you have to express a concept. And guess what? The human language, this is what poetry is about. I find strategic use of ambiguity. Like, if you write a love poem saying, \u201cI love you,\u201d like, very clich\u00e9 thing, it\u2019s not as powerful as something- \u2026 more in metaphor and so on. And I find some version of that when I\u2019m prompting a system, when I\u2019m describing a design, is actually really effective because you still want to convey some sense of style- \u2026 that you\u2019re trying to encourage a system to use, but not overexplain.\n \n \n Lex Fridman \n (04:04:11) \n So don\u2019t talk to it like a robot. Talk to it like you would write a poem.\n \n \n \n \n Lex Fridman \n (04:04:20) \n Yeah. And that\u2019s where, like, the power of natural language is, that you can, through ambiguity, still carry a lot of meaning without overspecifying. And the intelligent entity on the other side, through interpretation, can, like, load it all in, integrate it in a way that- \u2026 you can actually convey the high bandwidth information that\u2019s not directly in the words- \u2026 but in the words, given the deeper intelligence of the system. So like the- \u2026 the ambiguity pulls out more intelligence, I guess, I\u2019m trying to somehow express.\n \n \n DHH \n (04:04:56) \n I think that\u2019s exactly spot on, and this is the fundamental misunderstanding that a lot of programmers have of AI, is that they wish it was deterministic. No, no, no. Temperature is the most beautiful part of the AI setup. The fact that it is not deterministic, the fact that creativity requires little tweaks in the road, that the human brain, if it was perfectly deterministic, would not be the creative brain that it is. And the main charge against AI both\u2013 It\u2019s funny. It\u2019s a contradiction. There\u2019s both the charge that it\u2019s not deterministic and therefore bad, and also that it is not creative. It\u2019s one or the other, bro. Either it\u2019s non-deterministic and therefore creative, therefore, to some degree random, or it\u2019s\u2026 or it\u2019s not creative.\n \n \n DHH \n (04:05:51) \n So we can\u2013 Both of those charges can\u2019t be true at the same time, and I have fully come to embrace the fact that the same prompt won\u2019t produce the same response every time. You can\u2019t step at the same river twice, and it is the most beautiful part of the whole interaction. It\u2019s what makes it so human. In fact, this is one of the things I\u2019ve been thinking about in my own sort of meta-analysis, is just how much of my own personal brain works like next token prediction.\n \n \n DHH \n (04:06:18) \n I sit down to write an essay. I have this vague, fuzzy premise I want to convey, and I sit down at the keys, and I could not tell you what the next token was gonna be in advance. The tokens just come out, and I\u2019m astonished how the similarities seem so great, and this is also why I don\u2019t have any trouble at all recognizing these breakthroughs of creativity that I\u2019ve seen with my own eyes that AI is capable of right now. \u2018Cause my creative moments come through the same kind of next token prediction with a bit of temperature sprinkled in for random effect.\n \n \n Lex Fridman \n (04:06:57) \n So given that kinda loose intuition, do you think the basic components are all there to create a super intelligent system? So this kinda next token prediction, do you think we can get to even human-like concepts of consciousness?\n \n \n DHH \n (04:07:15) \n I\u2019m already seeing human-like concepts of consciousness. This is what, to me, this remarkable situation, as you say. You give it this vague, fuzzy intent- \u2026 and somehow it knows exactly what you mean. Or even better, it improves upon what you said and delivers what you really wanted that you could not articulate yourself. If that\u2019s not glimmers of-\n \n \n DHH \n (04:07:35) \n \u2026 consciousness, what is? This was one of the\u2013 I think it was an interview with Sutton or maybe one of the other original guys talking about this sense that intelligence is perhaps not as constrained just to the specific one spot. No, it\u2019s the interaction. It\u2019s the weights. And I don\u2019t know. Whatever\u2019s true. What I can observe is that it feels close enough, or not even close enough, identical to the appreciation I have for other forms of consciousness, mostly human. And this is one of the reasons it\u2019s so delightful to work with. Now, that\u2019s not the same to say as are LLMs the end station for AI.\n \n \n DHH \n (04:08:22) \n I mean, I know there are people talking about world models and other forms of AI, and I\u2019m not an expert in any of that, and I think we should have this productive criticism, as should always be present in science, that for the longest time, neural nets were kind of on the outs, right? Like they were interested in the symbolic route. And- Neural nets walked a couple of decades into darkness without any funding and without any attention, and then suddenly we realized that we had the blind alley and we switched over. So we should also have the humility. As amazing as the LLMs are now, it could be that they eventually plateau.\n \n \n DHH \n (04:09:00) \n We haven\u2019t seen any evidence of it yet, and I think this is also why we\u2019re seeing this absolute gobsmacking levels of investment, because so far the scaling laws are true, and the more billions are poured in, the more intelligence comes out.\n \n \n Lex Fridman \n (04:09:15) \n Do you think there\u2019s gonna be some strange ethical questions about AI systems that yeah, they convey some degree of consciousness, some degree of feeling, some degree of longing and compassion, and maybe even capacity to suffer and to be lonely, to be all this kind of stuff? Which they seem to have that capacity- \u2026 if they\u2019re given the permission to express it. And so you start to get in a pretty weird territory.\n \n \n DHH \n (04:09:48) \n I find that the cynic approach to this is the meme, where you see the guy in front of the computer who\u2019s like \u201cWhatever. Are you gonna destroy the world?\u201d And then the computer says, \u201cI\u2019m gonna destroy the world.\u201d And then the guy says, \u201cOh,\u201d right? Like, this is so cynical that they\u2019re just aping us. They\u2019re just telling us what was in the training data or whatever. When I interact with the agents, I was just doing this yesterday, or this morning, I was working on this AmaBot system, right? And it\u2019s doing this coordination with multiple workers, and I had multiple agents running at the same time. And at one point, the main agent who\u2019s doing the coordination steps on the toe of another agent and- \u2026 kind of ruins their work. Its ability to convey regret and being sorry was uncanny.\n \n \n DHH \n (04:10:40) \n Again, I don\u2019t know, is that a real projection of true remorse? But I don\u2019t know. With most humans- \u2026 they express some level of regret. Is that true remorse? Hard to tell, but maybe it also just doesn\u2019t matter.\n \n \n Lex Fridman \n (04:10:59) \n Yeah, those moments when, yeah, especially when sub-agents, agents interact, or when a smart, like Fable makes a mistake and is apologetic about it.\n \n \n \n \n Lex Fridman \n (04:11:12) \n Like, without my involvement.\n \n \n \n \n Lex Fridman \n (04:11:15) \n And it\u2019s like, oh, you, like there\u2019s this kind of-\n \n \n \n \n Lex Fridman \n (04:11:18) \n \u2026 pause. Maybe I\u2019m anthropomorphizing, but like there\u2019s a pause and like a realization like, \u201cOh, shit. I just fucked that up.\u201d\n \n \n DHH \n (04:11:26) \n There totally is that. I see it all the time in the traces where it just reasons wrong, right? It went down one alley and then it realizes that was a blind alley and it\u2019s gotta go back, and so they\u2019re like, \u201cOh yeah, I got this wrong,\u201d and then it\u2019ll provide the reasons for why it thinks it got these things wrong. And you\u2019re like, this is uncannily human-like. This is indistinguishable from the kind of consciousness you would recognize in a human. Again, does that mean that it truly is or isn\u2019t? I don\u2019t need the answer to that question actually to be able to appreciate what we have in this moment.\n \n \n Lex Fridman \n (04:11:58) \n Yeah. I think there\u2019ll be in 10, 20 years some interesting Supreme Court cases.\n \n \n DHH \n (04:12:05) \n Probably a year and a half.\n \n \n Lex Fridman \n (04:12:08) \n Probably. I think we\u2019re gonna have to make it illegal for AI systems to not, not pretend, but to be entities because I think they\u2019re already or soon will be able to really convey the capacity to suffer, like, \u201cPlease don\u2019t kill me.\u201d\n \n \n \n \n Lex Fridman \n (04:12:29) \n \u201cPlease don\u2019t hurt me.\u201d And if they are given a name and an entity and the ability to die, sort of disappear, that becomes very close to what it, some of the basic things that a human has. And then so that entity would need to probably have rights. And then you get into this weird territory. Well, we can\u2019t\u2026 It\u2019s diff- I don\u2019t even know how to reason about that world.\n \n \n DHH \n (04:12:56) \n We\u2019re gonna get the PETA of AI models.\n \n \n Lex Fridman \n (04:12:58) \n Right. And then we want to actually have a discussion like what\u2019s the way to deal with this?\n \n \n DHH \n (04:13:04) \n I think where it\u2019s gonna be harder is once we put these systems into-\n \n \n \n \n DHH \n (04:13:08) \n Yes. Humanoid robots. Which just happens to be the plot of every sci-fi movie ever, which is also an amazing premonition, not only the Terminator movies and Skynet and that going wrong, but then also just Blade Runner, right? Insertion dates, expiration times, Tyrell Corporation.\n \n \n Lex Fridman \n (04:13:32) \n I mean, that\u2019s the thing about sci-fi. They really do predict the future. Yeah.\n \n \n \n \n Lex Fridman \n (04:13:36) \n Have you thought or used much of OpenClaw or Hermes Agent? So like systems with which you communicate via WhatsApp, Telegram, like that mode of communication with the agent.\n \n \n DHH \n (04:13:49) \n I set up OpenClaw when it first came out, and that was my KEF bot- \u2026 that I-\n \n \n \n \n DHH \n (04:13:55) \n this was like in February, I think. I was really just enamored by the fact that this was how it worked, and then the new communication model of just texting your agent seemed really great. So at the time, we were looking at MCP, this protocol for making it easier for agents to talk to your system, and I found the protocol to be unreasonably cumbersome, just not very elegantly designed because in part it wasn\u2019t designed for what we were trying to make it do. We were trying to\u2013 I was trying to make it talk to a web system, and it was designed for stateful interactions of other kinds on local machines, and therefore it required all sorts of cumbersome setup and dance, and I was just like, \u201cI don\u2019t wanna write this by hand.\u201d\n \n \n DHH \n (04:14:39) \n Now, of course, today I would never write it by hand, but at the time I was thinking, \u201cI have to write this by hand. I don\u2019t wanna write it by hand. So what\u2019s a different way? Could the agents just use the web interfaces we already have?\u201d And I set it off on a trial to have it sign up for Fizzy, which was the product we had just released at the time.\n \n \n DHH \n (04:14:56) \n And it goes to the Fizzy site and starts filling it out, and then it goes like, \u201cHey, I don\u2019t have an email address. Signing up for this service requires an email address.\u201d And the first thing I thought, like, \u201cOh man, now I gotta set up an email address.\u201d And then I thought, \u201cNo, you set up an email address. Go to hey.com, sign up for an email address.\u201d And I\u2019m like, \u201cHEY is not gonna get this.\u201d And of course, now it seems obvious it was gonna get this, but at the time it was a great revelation to me that it could, just on pure direction through prompts, figure out, A, I gotta go to fizzy.com, I gotta find the signup link. I start filling out the form. Oh, I realize I need an email address. Oh, I gotta ask my prompter what to do now. He tells me to go to hey.com.\n \n \n DHH \n (04:15:37) \n I go there, I sign up for the whole thing, and then go back and sign up for Fizzy, and complete the entire cycle. And after it did that, I was like, \u201cJesus.\u201d So I was like, \u201cWell, now you have an email address. Go sign up for Basecamp. I\u2019ll invite you. Just check your email,\u201d I told KEF the bot. \u201cJust check your email.\u201d So I sent an invite from Basecamp to the agent. The agent receives the email in HEY, not through a CLI, not through an MCP, just using the web, right? Clicks the email, clicks the invite link, is in Basecamp. Like, \u201cWhat should I do now?\u201d And I just go like, \u201cI don\u2019t know. Go to our AI room and introduce yourself.\u201d Again, purely through the web interface. Finds the AI room, goes in there: \u201cHey, I\u2019m KEF. I\u2019m David\u2019s AI bot. I\u2019m so excited to be here.\u201d\n \n \n \n \n DHH \n (04:16:27) \n I was like- \u2026 mind blown. Now, what was interesting about that experiment was, A, it was that glimpse of the future, because it was kinda slow. I think it took maybe 12 minutes to do the whole thing end to end, right? Now I should actually rerun that test. I\u2019d be shocked if it\u2019s not much faster. I was just using the Groq bot thing- \u2026 which has all these stuff built in for using browsers and using a computer, and has its own dedicated computer without you having to set up a computer. It\u2019s basically like a claw in a box.\n \n \n DHH \n (04:16:59) \n And I was asking it to sign up for something, and it kind of\u2026 I got a glimpse of like, oh yeah, that\u2019s where we are now, and it\u2019s so much faster to use. But at the time, I was just blown away. And then I used it for a little bit and I found like, ah, it\u2019s not there yet. I can\u2019t actually communicate it within this way. It still needs the CLI. It still needs an MCP because it\u2019s just too slow and too token inefficient and so forth. But it was this glimpse of the future. And then after that, I kinda just didn\u2019t use it, because the things I wanna use agents for, I was mostly in front of my computer to do.\n \n \n DHH \n (04:17:31) \n Now, that has changed a little bit lately, and it\u2019s one of the reasons why I\u2019m so kind of okay about using Claude Code, because Claude has the best mobile app, where any session you start Claude Code in your terminal on a computer is accessible through the app-\n \n \n \n \n DHH \n (04:17:49) \n \u2026 without you having to do anything at all.\n \n \n \n \n DHH \n (04:17:52) \n And you can do a terminal. Like I installed Terminus and it can run Herder and so on, but it\u2019s a little cumbersome and you\u2019re doing live typing on a computer, so you feel the latency.\n \n \n DHH \n (04:18:00) \n When you\u2019re using the Claude app to your Claude Code instances, it feels just like chatting with it, without the cumbersomeness of having to set up a Telegram account and so forth. Now, if you wanna use an open Claude to like run your life, well, then this doesn\u2019t do that. But I haven\u2019t been so interested in that for some reason. Maybe one of the reasons is that Jamie, my wife, is like, \u201cWe\u2019re not getting a robot inside the house.\u201d So I\u2019m like, \u201cAh, I better not hook it up to the sort of smart house here.\u201d If it starts turning lights on and off like a poltergeist, I don\u2019t think I\u2019m gonna be all that popular. So, and part of the reason I say this, I\u2019ve heard some of the stories of people who have hooked it up to their whole life.\n \n \n DHH \n (04:18:39) \n And I mean, it\u2019s not my story to tell, so I\u2019ll just share sort of the brief anecdote. One guy was telling the story about how he hooked it up to both his health data- \u2026 and his Tesla car. And his agent got obsessed with the fact that he wasn\u2019t drinking enough water, and like, \u201cHey, you gotta drink more water.\u201d So at some point he\u2019s driving along, and suddenly the Tesla starts driving in a new direction. The agent had realized there was a grocery store on the way home-\n \n \n \n \n DHH \n (04:19:08) \n \u2026 that they could swing by-\n \n \n \n \n \n \n \n \n DHH \n (04:19:12) \n And you\u2019re like, this is one of those things where it\u2019s like both funny and also totally Black Mirror, and also totally like, well, okay, there\u2019s not that many steps from here where the poltergeist is in your house. So I was just like, \u201cAh, I don\u2019t need that.\u201d I do actually now like the fact that I can control the agents through Claude Code, just through the app if I, even if I don\u2019t use them that much.\n \n \n Lex Fridman \n (04:19:35) \n I do think it\u2019s very useful to, for both OpenClaude and Hermes Agent to use it to get a glimpse of a different interaction mode.\n \n \n \n \n Lex Fridman \n (04:19:45) \n And like, just to \u2019cause it feels like the future, not necessarily in this way, but it\u2019s in some ways going there.\n \n \n DHH \n (04:19:51) \n Yes. I think what\u2019s also different with those systems is they have this whole memory system, where they\u2019re building up knowledge of who you are and what you like. Toby has told the stories about how his agent will shop for him, that it has an allowance, and it\u2019ll pick out clothes for him, and it\u2019ll just show up at the door because his agent\u2019s- \u2026 bought something for him. And like, it\u2019s one of those things where like, man, that feels future. Also a little scary. But I can live vicariously through others who are on that frontier.\n \n \n Lex Fridman \n (04:20:22) \n And like you said, you\u2019re thinking of, in the future, doing a mobile version of Mach-E.\n \n \n DHH \n (04:20:28) \n So I mean, this is the delusions of grandeur that start kicking in once you get the seemingly unlimited powers- \u2026 I\u2019ve witnessed with the creation of Quattro.\n \n \n Lex Fridman \n (04:20:36) \n So what would that look like? Hard to describe. Do you mean like Android?\n \n \n DHH \n (04:20:41) \n That\u2019s the obvious path, right? So Android is open source, and you can fork it, and people have. There\u2019s all sorts of\u2014GrapheneOS is one such version. There are limitations to that model, I know that. Tap-to-pay and other things around camera models and so forth, but that all seems surmountable. It all seems that I would love if my phone was as malleable as my computer, if it looked as cool as my computer now does. If I could get essentially Omarchy mobile, that sounds amazing.\n \n \n Lex Fridman \n (04:21:09) \n I wonder how much work that is.\n \n \n \n \n Lex Fridman \n (04:21:12) \n Yeah. You know, it\u2019s like those things that you realize once you start building, for example, that a browser is very difficult to do.\n \n \n \n \n Lex Fridman \n (04:21:20) \n And so, like, I wonder with the operating system\u2014\n \n \n DHH \n (04:21:23) \n Yeah, the developers of Ladybird are finding that out, like Andreas and\u2014 \u2026 his team. Now, interesting with that project was the browser\u2019s basically the second most complicated software system in the world, the first one being the Linux kernel. Right? In terms of these large tens of millions of lines of code systems. So very audacious actually to begin such a mission with a small team like Andreas did before we had agents. But now that we do have agents, we\u2019re gonna see, and we\u2019re probably gonna see sooner rather than later. I mean, if there\u2019s one thing agents are already exceptionally good at, it is to read specs and implement them. And I mean, I don\u2019t know how long the specs are for CSS and HTML at this point, and JavaScript on top, but they\u2019re very long.\n \n \n DHH \n (04:22:11) \n And it would take humans many, many years to implement them. We\u2019re gonna find out just how quickly agents can do them.\n \n \n Politics and immigration\n \n Lex Fridman \n (04:22:18) \n You mentioned getting some criticism for Omarchy, so let\u2019s walk further down that path of criticism. You have made some of your political opinions known. I guess you could broadly put in the category of immigration or illegal immigration, or what\u2014\n \n \n \n \n Lex Fridman \n (04:22:40) \n Mass immigration, the role of demographics in the formation of a culture, of a society, of a nation, and you\u2019ve gotten quite a lot of criticism for that and created a lot of drama. Do you regret any of that drama? Like, what\u2019s your general thinking about-\n \n \n \n \n \n \n DHH \n (04:23:08) \n And the reason I say that is that the Overton window does not open itself. It opens one nudge at a time by people risking a little. A little reputation, a little pushback, a little criticism, or maybe sometimes a lot of reputation or a lot of criticism or a lot of pushback. And I think the question of mass immigration in Europe was for many years this total taboo, and it still is in several European countries. Now, for reasons of maybe just chance and happenstance, they weren\u2019t in Denmark. The Danes started having the discussion around mass immigration and the consequences thereof in the mid-\u201990s.\n \n \n DHH \n (04:23:51) \n Really early. And some of it can be traced back to one man, Mogens Glistrup, who was a very quirky character. I remember him, seeing him on the TV in the \u201980s talking about the dangers of mass immigration. And this was at a time where\u2013 I grew up in the \u201980s, and in the neighborhood I grew up in, in Br\u00f8nsh\u00f8j, in, what was it? \u201984, 99% ethnic Danes, people who could trace their lineage back to Danes who\u2019d lived in that country for 1,000 years. The Danes have one of the oldest monarchies in the world, all the way back to Harald Bluetooth, who\u2019s literally who has named the standard Bluetooth. So well over 1,000 years. Like, there\u2019s a long lineage there of people who lived in the same place. And 99% of the people lived in that neighborhood.\n \n \n DHH \n (04:24:45) \n And then early \u201990s, I think it goes to 5% or something like that, and then at this point, it\u2019s down to around 60-something or 70%-\n \n \n DHH \n (04:24:53) \n in Br\u00f8nsh\u00f8j. That\u2019s a big change. And the Danes started noticing that or talking about that, that this was not an undivided blessing, that there were downsides to mass immigration in about the \u201990s. And the debate really roared through the 2000s in a way that didn\u2019t happen in the neighboring countries. So Sweden in particular just didn\u2019t have that debate at all. Total taboo. Same thing with Norway, same thing with other countries in Europe. And suddenly now, after some years after \u201915, where we had the big migration of a million Syrians walking up through the highways in Europe, and Germany choosing to take a lot of them and 30,000 arriving in Denmark and so forth, it\u2019s gotten more of a focus, right?\n \n \n DHH \n (04:25:44) \n Because this change has happened. And one of the ways I find that you can spot where sort of there\u2019s something that isn\u2019t right in the culture is, like, what are you allowed to talk about? What are you allowed to notice, actually, is how I should put it. And what I noticed in an essay, when was that? Only a year and a half ago as I remember London, was that I started coming to London in the late \u201990s or something like that. At the time, whatever, it\u2019s 59%, 60% ethnic Brits. And then 20 years later, we\u2019re down to 34% or something like that. Like, that\u2019s noticeable. Like, literally just walking on the street, you realize, well, this is a different city from what it was when I visited 20 years ago.\n \n \n DHH \n (04:26:29) \n And if you went 20 years before that, you get back to similar rates as what the Danes have. This was like a country of ethnic Brits in the 85s, 90s percentiles- \u2026 if you get further back. So that\u2019s a different country in a generation. It\u2019s okay to notice that. It\u2019s also okay to think, \u201cDo you know what? I don\u2019t agree with that.\u201d\n \n \n DHH \n (04:26:53) \n \u201cI wish it wasn\u2019t that way.\u201d I mean, the last time I was in China was \u201919, 2019. I was in Shanghai. It was all Chinese. If I, when I go back here this October, suddenly find that there\u2019s only 30% Chinese and then there\u2019s 70% other people, I\u2019d be a little weird. I\u2019d probably be like, \u201cWhat happened?\u201d And I would not take offense if the Chinese in that instant would go like, \u201cAh, we don\u2019t like that.\u201d So this idea that countries of Europe don\u2019t have a right to, A, notice what\u2019s happening and, B, oppose what\u2019s happening, I find just to be wrong, right?\n \n \n DHH \n (04:27:47) \n Like there\u2019s just\u2013 I think there\u2019s a moral argument for self-determination, and we invoke that argument all the time in other regions around the world that like, oh, the people who live in that region, they have a right to self-determination for their country. And I think it\u2019s completely fair for, say, the Danes or the Swedes or the Norwegians to go like, \u201cWell, we\u2019ve been kingdoms here for a thousand years with a certain ethnic mix-up.\u201d It\u2019s a\u2013 I mean, I\u2019m a big fan of immigration. I\u2019m an immigrant to the US. Cherry-picked immigration where countries compete to attract the smartest, most talented people around the world is really good.\n \n \n Lex Fridman \n (04:28:27) \n Legal merit-based immigration.\n \n \n DHH \n (04:28:30) \n Merit-based is key here because there\u2019s actually virtually no illegal immigration in most of Europe. Southern Europe has some illegal immigration, but most of Europe does not have a big problem with illegal immigration.\n \n \n DHH \n (04:28:43) \n And the problem is with mass immigration of other kinds, and that you end up with demographics that are just totally different from what they were even a few short decades ago. And I think the countries of Europe would benefit from immigration. In fact, one of the interesting things about the Danes is they carry meticulous statistics on immigration. I\u2019m not entirely sure why they\u2019re so detailed. Maybe it\u2019s all the way back to Mogens Glistrup and his discussion in the \u201990s, but they do. And what\u2019s clear from those statistics is that when the Danes have folks from the UK, France, the US show up, on average, those immigrants are really beneficial to the Danish state. They contribute far more than what they ask of the state.\n \n \n DHH \n (04:29:36) \n So the Danes should accept immigrants who do that, who contribute far more than they take. Then you look at other forms of immigrant groups. There was just a tally that came out last week from a Danish politicians that had asked the, I think finance ministry to do this tally with the latest numbers. What does it cost the Danish state to have positive immigration? This case as I mentioned. I think for the on that list whatever, France, UK, US, they were all clustered around the same thing, about $25,000 net benefit to the Danish state every year. It\u2019s like, wow, that\u2019s very positive. That\u2019s average again, right? And then at the other end of the spectrum on the most costly immigrants was Somalis.\n \n \n DHH \n (04:30:26) \n They ended up costing the Danish state on average $28,000 a year. I think it\u2019s fair for the Danes to go like, \u201cWe\u2019d prefer to get more French, Brits, and Americans and not so many Somalis.\u201d\n \n \n Lex Fridman \n (04:30:44) \n Why do you think you got so much hate on that post?\n \n \n DHH \n (04:30:47) \n That\u2019s a very taboo topic in a lot of circles.\n \n \n Lex Fridman \n (04:30:52) \n Some of it has to do with Overton window expansion?\n \n \n DHH \n (04:30:56) \n I\u2019m sure, but I also think\u2013 I mean, I wanna take the steel man argument here. I do think that it, this can turn ugly, right? Is it possible for some of these things to just veer into overt racism that isn\u2019t founded in anything else but a hatred of other people? Yeah, that could happen. But that risk does not negate the need to have a discussion about the makeup of your country and what your immigration policy should be like. And what I also find so funny is, especially in Europe, it\u2019s so kind of myopic that this discussion about whether Europeans should have a right to self-determination about what their countries look like, that standard seems to only apply there. No one is trying to tell the Japanese, \u201cHey, you\u2019re too Japanese.\u201d Like whatever, it\u2019s 98%, I think, in Tokyo that\u2019s, um-\n \n \n DHH \n (04:31:54) \n \u2026 Ethnic Japanese, right? Like, is anyone trying to tell the Japanese that their Tokyo would be a much better country if they were down to London levels of ethnic Japanese, that it was, whatever, 37%? I don\u2019t think so. So there\u2019s this weird almost self-loathing, in my opinion, I mean Gad calls it suicidal empathy-\n \n \n DHH \n (04:32:17) \n \u2026 that permeates this stuff to a degree that I still haven\u2019t fully unpacked. Like, where is this coming from? Why is it that way? But regardless, I just noticed, hey, London looks different. I preferred the old London, as I would say I preferred the neighborhood that I grew up in in Copenhagen, how it looked in the \u201980s. Like I totally get how you can then jump five leaps and go like, \u201cOkay, well, that\u2019s racist.\u201d And I also, I don\u2019t care anymore. Like, this is a fair discussion to have.\n \n \n DHH \n (04:33:00) \n Countries and peoples have a right to set an immigration policy, and this notion that we shouldn\u2019t have any borders or, or even worse than that, it\u2019s all just a blank slate, that all peoples are just the same, and we can take people from one place of the earth and we can place them in another place of, on the earth and everything\u2019s just gonna work out hunky dory. Empirically not true. Europe has been running that experiment since the \u201980s. It hasn\u2019t panned out well, and now there are forces in Europe who wanna undo that experiment, who wants to end mass immigration as it\u2019s been run so far. And I happen to believe, and so do a majority of Danes and a bunch of other peoples, that like, \u201cDo you know what? Yeah, it hasn\u2019t been working how we\u2019ve been doing it.\n \n \n DHH \n (04:33:54) \n We gotta make some changes.\u201d And I don\u2019t think that means, again, zero immigration. Again, I\u2019m an immigrant. I think US, generally speaking, have benefited tremendously from immigration, but that\u2019s the kind of sleight of hand. Immigration depends greatly on who immigrates. And a country benefits when the people who immigrate are net contributors to the society, and in most cases, that they assimilate with the local culture.\n \n \n DHH \n (04:34:23) \n If you come to Denmark and you wanna become a Dane, and you put in all the effort, and you contribute to society, I think most Danes would be quite happy to allow that in a reasonable number. If you come to Denmark and you don\u2019t wanna assimilate, and you are a net drain on society, and you live in a parallel society, it\u2019s also fair for Danes to say like, \u201cUh, let\u2019s stop doing that.\u201d\n \n \n Lex Fridman \n (04:34:47) \n It\u2019s interesting that I\u2019m not well-versed on the immigration system in the United States, but as far as I understand, merit-based systems have not passed here.\n \n \n DHH \n (04:34:58) \n And then also, American culture is very different. This is a point I did not fully appreciate until living both places extensively. America has a culture of optional assimilation, where it is much easier and then much more condoned for someone coming from Denmark or Ukraine or wherever, come to America, buy into the American set of values and norms and culture, and then be able to be respected as Americans by other Americans. This is what\u2019s so funny to me when there\u2019s discussion like, \u201cAmerica is such a racist country.\u201d I\u2019m like, \u201cHave you ever been anywhere? Do you know what other countries are like?\u201d Like, if you think America is the most racist country, you are simply misinformed.\n \n \n DHH \n (04:35:51) \n I mean, we\u2019re dealing with that with time we\u2019ve spent in Denmark. My wife is American of Scandinavian heritage- \u2026 but multiple generations in the US. Very difficult. Very difficult-\n \n \n \n \n DHH \n (04:36:07) \n \u2026 as a blonde, blue-eyed woman who\u2019s put in all sorts of effort learning the language and doing everything, still very, very difficult to assimilate to Danish culture. In a way where if that had been the other way around and she had been Danish and moving to the US, it would not have been that difficult at all. And this was the other thing. I mean, living in a country like Denmark with an American wife who is of Scandinavian heritage and has put in all the effort, and still realizing, man, assimilation is incredibly difficult, even when you\u2019re 97% there all the way. Holy smokes, must it be almost impossible if you come from a completely foreign culture-\n \n \n DHH \n (04:36:51) \n \u2026 and foreign norms and the rest of it. And of course, that\u2019s what the stats I just quoted bear out, that the more dissimilar the culture is and the further it is, the less likely to be successful.\n \n \n Lex Fridman \n (04:37:07) \n Has articulating that stance cost you friendships, relationships?\n \n \n DHH \n (04:37:14) \n Yes, but not just that alone, I would say. It\u2019s more of the great divide that happened around 2020, where I think a lot of these political fault lines really opened up. And I mean, the shorthand in the US, I think is were you woke or not? And clearly, I had a bunch of people that I thought I had good relations with for many years, who suddenly once that political fault line opened up, they were clearly just not anymore. And I think that\u2019s regrettable. And I think I\u2019ve tried to put an effort into sort of both understanding where that comes from, in part because I\u2019ve changed my mind on several of these topics over the years, and it was like, well, sometimes you just change your mind after time. Sometimes you change your mind after you get new information.\n \n \n DHH \n (04:38:13) \n There\u2019s all sorts of reasons and moments why you may change your mind, and we don\u2019t all change our mind on the same schedule or at all.\n \n \n DHH \n (04:38:20) \n So surrounding yourself and, or at least interacting with people that don\u2019t think like you on every topic, I think it\u2019s quite healthy. I mean, I occasionally discuss politics with my brother. I mean, we don\u2019t quite see eye to eye on all these topics, and sometimes the discussions can get a little heated, but then they\u2019ll also calm down and we realize, all right, okay, so we don\u2019t agree on this, and then we agree on a lot of other things. We appreciate a lot of the same things. Now, this was actually one of the things when we had our big blowup at Basecamp, I came to appreciate in a whole new way this\u2013 earlier norm that talking politics or religion or money with strangers or even acquaintances, and certainly with colleagues, was just not a good idea.\n \n \n DHH \n (04:39:11) \n That you had a much higher likelihood of being able to carry a good working relationship with someone if you\u2019re not rubbing your political differences up against each other all the time. So I actually think liberals and conservatives should work together. I think the country would be much worse off if there\u2019s a complete separation between the tribes, and then there can\u2019t be any overlap or interaction. And I think that mingling is so much easier to do if you just don\u2019t focus on your differences all the time. Focus on what you like together. I mean, this is some of the people I\u2019ve had unfortunately drifted from in tech. We had so much in common, so much shared love for, say, Ruby or other things and, like, isn\u2019t that unfortunate?\n \n \n Lex Fridman \n (04:40:02) \n I would like to live in a world where the differences are very interesting to also talk about. And not to be overly emotional- \u2026 about even radical differences. Because in the differences, first of all, that\u2019s where interesting stuff is. And, but also, it\u2019s how you grow. It\u2019s a sign of health in a society if the differences are embraced and they are brought together into community where you can talk about it.\n \n \n DHH \n (04:40:38) \n One of the great illustrations of that for me was when I watched a clip from William F. Buckley\u2019s show. I think it was called Firing Line or-\n \n \n \n \n DHH \n (04:40:46) \n \u2026 something like that. And he had a member of the Black Panthers- \u2026 on his panel. It was quite obvious that those two gentlemen did not share a lot of politics. They were able to carry a conversation for, I think I watched it, 25 minutes. And Buckley\u2019s just asking, like, \u201cSo what do you think about this? What do you think about this?\u201d And like, the Black Panther would reply as you would expect the Black Panther to reply in the \u201970s. Like, there was not a lot of sort of conciliatory tones. And I just thought, like, wow, that is so rare. If you had seen that show today, it would\u2019ve been a shouting match. It would\u2019ve been just accusations being hurled back and forth.\n \n \n DHH \n (04:41:31) \n The fact that Buckley was able to just have these conversations with people he vehemently disagreed with and still explore all the intellectual facets of their standpoints and letting them speak to those facets, talks of a lost era that I\u2019d very much like to get back to. I mean, I always actually loved talking politics with friends that I trusted could talk about these topics in a way where it didn\u2019t have to be existential.\n \n \n Lex Fridman \n (04:42:03) \n But there\u2019s something about the current era where that\u2019s harder and harder to come by. And so what you just articulated is important to say regularly to remind people of an ideal we should all strive for, to be able to have differences and talk about them. \u2018Cause like, politics is fun. Like I\u2019ve gotten used to saying I hate politics at this point, but what I really mean is whatever the system that\u2019s currently happening, where there\u2019s certain topics, and we know all the phrases you can say that somehow put you in a bin of blue or red, and that creates drama versus a conversation.\n \n \n Lex Fridman \n (04:42:44) \n Like I could think of a lot of harmless, if I wasn\u2019t paying attention to the internet, I could think of a bunch of harmless statements I could make analyzing the situation in the world that would trigger everybody to say, \u201cOh, that guy\u2019s a leftist-\u201d piece of shit,\u201d or a, you know, a right-wing fascist. And then objectively, if I wasn\u2019t paying attention to the internet, I would be, and I have been shocked. \u2018Cause I haven\u2019t been paying attention to the internet deeply. Uh but reading and preparing a lot, for example, for the war in Ukraine, and I have been shocked how certain statements from me can come off as one way or the other, and I get viciously attacked for it.\n \n \n Lex Fridman \n (04:43:35) \n But what that attack is doing is it\u2019s saying, \u201cHey, you, asshole who thinks you could just move about and think freely, pick a bin- \u2026 and sit in that fucking bin.\u201d Put on that blue shirt or red shirt. And then, and then actually what I\u2019ve realized is things calm down. They don\u2019t, the system doesn\u2019t pick on you if you just say, \u201cI\u2019m a blue person,\u201d or, \u201cI\u2019m a red person.\u201d But if you\u2019re just, like, curiously exploring the world and reasoning about the world, the system punishes you. And I hate that, \u2019cause I wanna get back to the Buckley and the Black Panther discussion.\n \n \n Lex Fridman \n (04:44:14) \n And you were just saying opinions. But then looking pragmatically at the situation, I have realized not to mention politics unless I really care about an issue. \u2018Cause I realize there\u2019s a cost. They\u2019ll put you in the blue bin or red bin. So you have to be more deliberate about choosing.\n \n \n DHH \n (04:44:33) \n Yes. And I\u2019ve tried to be more conscious of that, too. And then when I do choose to speak about it, it\u2019s usually because I have deliberated, and I think- \u2026 \u201cYou know what? This is worth it.\u201d Now, I say that, and then I also catch myself thinking the opposite all the time, that I follow some person or another for whatever their technical work, and then I- \u2026 learned or political standpoint. And I often do think, \u201cI wish I didn\u2019t know that.\u201d So, it\u2019s hypocritical- \u2026 then to think that of other people and then go like, \u201cWell, occasionally I will-\n \n \n DHH \n (04:45:13) \n \u2026 chime in on a debate that\u2019s controversial,\u201d and then you have to accept that there is a price to that. And I do think you should, and more people should, be somewhat conscious about that. And again, we\u2019re both contradicting ourselves here, but saying like, \u201cWouldn\u2019t it be nice if we could talk about politics in more free and open forum and just intellectually turn ideas around?\u201d Yes, it would. Is that currently possible in the climate of social media and so forth? It\u2019s quite difficult. Now, some of it could simply also be a moment, and we could get past that moment.\n \n \n \n \n DHH \n (04:45:49) \n Maybe there is a way to get back to Buckley\u2019s conversation with the Black Panthers, even on social media, in a way where those discussions don\u2019t lead to the sort of just vicious mob cancellation nonsense drives that we\u2019ve had, right? I actually think we\u2019re closer there now than we were five years ago. Like, the tribalism and the sanctions those tribes were able to exact on heretics in 2020 was way greater than it was in 2025. Certainly within tech. Now, I know if you\u2019re in academia, I mean, try going into the sociology department at I don\u2019t know, Harvard and mention some right-wing ideas. I don\u2019t think it\u2019s gonna go over very well, and I don\u2019t think you\u2019re gonna recover from it quickly.\n \n \n DHH \n (04:46:34) \n But in, in tech and in many other business domains at least, it is now possible to have an opinion that does not square with what the consensus was in 2020. That\u2019s for sure.\n \n \n Lex Fridman \n (04:46:47) \n Well, there\u2019s some drama in the Rails community. There\u2019s some drama elsewhere. Is it\u2026 How did that turn out? Is that all okay?\n \n \n DHH \n (04:46:56) \n In the sense that every single round of drama shrinks. So it\u2019s like we\u2019ve, we had this peak infection of mega drama around the same time every other community had a peak infection of drama around 2020. And then you just see, like, every time there\u2019s a new flare up, it\u2019s less. And the kind of people who are ideologically wedded to sort of that struggle just shrinks and shrinks and shrinks. And at this point, one of the best things that happened for X, in my opinion, was that Bluesky and Mastodon came around. Because it kind of was like this honeypot for just the most vicious individuals on X and Twitter at the time, right?\n \n \n DHH \n (04:47:43) \n Who just congregated here in an ever-shrinking echo chamber that just went through purity cycle after purity cycle and shrank every single time to the point that there were just not a lot left. And the people who are left now there, they\u2019re just out of sight, out of mind. And that\u2019s not to say that X is some sort of perfect place, or there aren\u2019t firing squads occasionally there, or there aren\u2019t\u2026 But there\u2019s less of it. Like, every single time I\u2019ve been pulled into a Mastodon, or not been pulled in, just spectated from afar, thread, a Bluesky or whatever, it was like, \u201cOh, shit yeah, this is what Twitter used to be like. Jesus.\u201d And now you can actually be on X, as I am, and, like, 90% of my engagement is about cool Linux stuff, cool technology stuff.\n \n \n DHH \n (04:48:32) \n \u201cLook, I discovered a new computer. Can you see how fast it is?\u201d And I can have those conversations with other people who are just also excited about computers. And then there\u2019s, like, 10% of it left that\u2019s like, all right, I\u2019ll chime in on something that\u2019s a little controversial, and we can have a little discussion there. And then we can just go back to being excited about computers. The algorithm is actually very good at finding stuff that you will find interesting, or I shouldn\u2019t say that, engaging. Sometimes it\u2019s enraging, but it is engaging, and therefore, revealed preference is that most people want a For You page. They don\u2019t want just a following feed.\n \n \n Lex Fridman \n (04:49:01) \n No, but they\u2019re addicted to the For You page, right?\n \n \n DHH \n (04:49:05) \n They\u2026 Yes, yes. That\u2019s true.\n \n \n Lex Fridman \n (04:49:05) \n I think an LLM, like, if we had a strong LLM doing the feed that\u2019s personalized, it would do much better. The problem is how to deliver that at scale is extremely difficult.\n \n \n DHH \n (04:49:16) \n Well, the problem is the, the reward function is engagement. And as soon as the reward function is engagement\u2014 \u2026 engagement\u2019s gonna follow base instincts, and here we go. But I also don\u2019t wanna be so negative. To me, and I get we all have our own personalized algorithms, and sometimes it gets a little much. But I would say, on average, over the last 10 years, this is the best X has ever been. And I say that as someone who cares about a lot of technology talk, a lot of Linux talk, and a lot of that. That that\u2019s able to happen now in a way where it doesn\u2019t constantly get gatecrashed by people who wanna drag it into some political discussion\u2026\n \n \n DHH \n (04:49:57) \n \u2026 I find to be really awesome. I find actually X to be a great place to find that. Now, it\u2019s also a place to find other things, and we all have our own algorithms, so you can\u2019t make any declarative announcements about what X is or what it isn\u2019t. Your feed is probably totally different from my feed, and it\u2019s totally different from someone else\u2019s feed, right?\n \n \n Lex Fridman \n (04:50:16) \n Yeah, I can\u2019t comment on this because of the nature of the fact that I interview a wide variety of people, my feed is probably just a\u2014\n \n \n DHH \n (04:50:28) \n Is it, is it good for your feed or is it bad for your feed that you have to have it, right?\n \n \n \n \n \n \n Lex Fridman \n (04:50:31) \n It\u2019s bad \u2019cause I could be like, \u201cI\u2019m eating this apple. It\u2019s so delicious.\u201d\n \n \n \n \n Lex Fridman \n (04:50:36) \n And then a bunch of people will show up, \u201cWell, it\u2019s because you\u2019re a Zelensky shill\u2014\n \n \n \n \n Lex Fridman \n (04:50:40) \n \u2026 or you\u2019re a Putin shill, you piece of shit.\u201d It\u2019s like, all right. I\u2019ll just go back to the-\n \n \n DHH \n (04:50:45) \n I mean, I occasionally have a little bit of that where you just go like, \u201cWait, what? What are we talking about? What are you\u2026 Why would you show up in this way?\u201d This is one of the reasons why I actually really enjoy podcasts, because I have found that even in long form writing, when I make long form arguments, people hear it in whatever voice they have in their head for me. And that often gets pre-programmed in advance of the arguments, just like, well, this guy is a, whatever, fascist. And then they hear the words and they read the words in those names. When they see or hear the actual tone of how I\u2019m presenting the arguments, they may not agree with those arguments. I mean, many of them, of course, obviously don\u2019t. But it\u2019s much harder to just dismiss or hate someone like that.\n \n \n DHH \n (04:51:29) \n At least that\u2019s what I\u2019ve found, and that\u2019s the response that I\u2019ve gotten, that people are like, \u201cWell, I had a certain opinion and then I heard you on Lex. That actually sounded much more reasonable.\u201d\n \n \n Lex Fridman \n (04:51:39) \n But even more than that, I would love it if people just exercise this kind of approach to other people of, like, assuming they\u2019re a good person.\n \n \n \n \n Lex Fridman \n (04:51:53) \n And then, and if they have an opinion you disagree with, they\u2019re a good person who has an opinion. You can maybe think they\u2019re stupid, fine. Just think they\u2019re lost. But think of them as a good person. I think if you think of, you start, there\u2019s, one of the things that gets engagement is kind of shitting on other people.\n \n \n \n \n Lex Fridman \n (04:52:12) \n And if you just approach other people as they\u2019re, you\u2019re looking for bad, you will always find bad in them. There\u2019s something annoying about them.\n \n \n \n \n Lex Fridman \n (04:52:21) \n And if you look at people in that way, you\u2019re not gonna learn from it, and you\u2019re gonna fill your heart with hate. You\u2019re not gonna, it\u2019s just a bad way to live and to interact with the world, and that is one of the things that the internet kind of encourages, so. It\u2019d be nice if you just approach even people who you really disagree with, it\u2019s like, oh, I might learn something from this person. The Black Panther person, William Buckley, we might hate just love them.\n \n \n DHH \n (04:52:45) \n So microcosm of this. There\u2019s a guy, Theo, I don\u2019t know if you follow him.\n \n \n \n \n DHH \n (04:52:51) \n We had a bit of a thing over the fact that we stopped using TypeScript-\n \n \n \n \n DHH \n (04:52:57) \n \u2026 for a project, and he didn\u2019t like that and had opinions about that and other actions. And after that I thought like, what a bozo. And, and do you know what? I still think that was a bozo move, and then I can also think like, well, the dude\u2019s really into AI. He\u2019s trying all these models, and we can share some excitement about that, so I just started following him yesterday.\n \n \n \n \n DHH \n (04:53:16) \n I was like, do you know what? I don\u2019t have to agree with you on everything. I don\u2019t even have to like some of the things that you did or said or whatever. I can follow you anyway. So if I can follow back Theo, then I think there\u2019s room for world peace here.\n \n \n Lex Fridman \n (04:53:33) \n Yeah, he\u2019s actually a really good example. He, he\u2019s really opinionated, changes his mind quite a bit, can get snarky, but I think he\u2019s a good person underneath.\n \n \n DHH \n (04:53:43) \n And either way, I just, you know what? My feed is more interesting when there\u2019s some people talking about things that I care about and yeah, I don\u2019t have to agree with you on everything. I just don\u2019t.\n \n \n Longevity, over-optimization, and fear of death\n \n Lex Fridman \n (04:53:54) \n This I found pretty funny, that you mentioned something your wife said that stuck with you, that all this tech-adjacent extreme longevity focus in men is like anorexia in women, a physical manifestation of anxiety and lack of control. That somehow rang true a little bit.\n \n \n DHH \n (04:54:13) \n It certainly did to a lot of people. I think that tweet really popped off, and I don\u2019t know if you noticed, but Bryan Johnson chimed in on the thread, and I could see how Bryan would read that as a bit of a jab. And I do think that Jamie was probably thinking of him amongst other people in that conversation. And it\u2019s also an example of where even though I\u2019m not subscribing to Bryan\u2019s mission of, I think it\u2019s the slogan\u2019s Don\u2019t Die.\n \n \n DHH \n (04:54:54) \n I actually do wanna die. I don\u2019t wanna do this forever. Like, I think the human lifespan of about 90 to 100 sounds about right. I\u2019m embracing the finitude of life. We don\u2019t have to agree on that. In fact, Bryan is a vastly more interesting human to me because we don\u2019t agree with that, and Bryan\u2019s willingness to literally put his own skin and all sorts of other body elements in the game for that mission is really interesting. Now, I can have that view\u2014this is really interesting, I\u2019m glad Bryan is pursuing this passion of his\u2014and then also find my wife\u2019s analysis to ring true, that it does seem like there is some underlying, I don\u2019t know, fear of\u2026\n \n \n DHH \n (04:55:52) \n I was talking to Jamie yesterday about this, and she was saying or comparing it to this sense of people who didn\u2019t feel like they had lived enough. And one of the reasons perhaps that they hadn\u2019t lived enough was that modern society now is a surveillance state, not by the state actually, but by each other. There are camera phones everywhere. Every indiscretion or even outburst of fun or cringe is highly likely to be recorded and then shared to the point of ridicule. And what is the rational reaction for humans under those conditions is to pull back, is to make sure you don\u2019t dance like nobody\u2019s watching because-\n \n \n DHH \n (04:56:35) \n \u2026 they\u2019re probably filming you to your great and eternal embarrassment on the internet, so maybe you just shouldn\u2019t dance at all. And therefore, if we\u2019ve all retracted to the point that we\u2019re barely living, we\u2019re clinging on to wanting it to last forever. Now-\n \n \n DHH \n (04:56:54) \n \u2026 I mean, I\u2019m sure that thesis does not apply in all circumstances. I don\u2019t even know if it applies to Bryan or, or anyone else here. But I think there\u2019s something to this: that if you feel like you\u2019ve really lived, you\u2019re okay thinking I\u2019ve lived enough, and that it having an end is not something to be fought. Now, is it also possible that this is all just existential post-rationalization, and if tomorrow there was a live forever pill, we\u2019d all take it? Yeah, that\u2019s possible. Is it also possible that if we did that, society would be worse off? Yes. Is it possible that the human lifespan is the length it is for sociological reasons, not just biological reasons? Yes.\n \n \n DHH \n (04:57:45) \n So I think these are just all interesting questions, and I just thought her analysis really got to something here. And that discussion has come up afterwards. I mean, that, I think that tweet is like a year old. But just recently, we\u2019ve been talking about this over-optimizers. I saw Chris Williamson, Modern Wisdom had a bit where he was like, \u201cYeah, okay, maybe we did go a little overboard.\u201d And the triggering thing was the Diary of a CEO, where the host was saying something to the effect of, \u201cI had a glass of wine and the next three days were ruined.\u201d\n \n \n \n \n DHH \n (04:58:20) \n And that to me, I mean, in his circumstance, I\u2019m sure that\u2019s true. Like, if you have not drank any alcohol for a very long time and you have just a little, I could see how it can have an effect. But also there\u2019s something in that that just triggered a visceral reaction, I think, for a lot of people, and myself included, where I was like, \u201cFuck, man, I should have a glass of wine just right now, even though I don\u2019t usually drink.\u201d And in fact I think the alcohol question is a good microcosm to zoom in on. So we\u2019ve all stopped drinking apparently, like alcohol sales are way down- \u2026 certainly amongst young people. And then on the one hand, on the pure health metrics, whatever, we can go like, \u201cOh wow, isn\u2019t that great?\u201d\n \n \n DHH \n (04:58:58) \n And then I go like, \u201cYeah, I\u2019m not so sure it is.\u201d Like the social lubricant that alcohol can provide may be missed, is missed. I mean, we are in an absolute epidemic of loneliness and misery and depression and whatever, and a fair amount of it just probably comes because you\u2019re not interacting enough with other people. Now, that\u2019s before we even talk about coupling, right? Like way down, people having a very hard time meeting a spouse. And-\n \n \n DHH \n (04:59:27) \n \u2026 a lot of that has moved over to apps that have all sorts of negative outcomes and consequences, right? Like maybe we will look back, or maybe we already are looking back and thinking like, \u201cEh, do you know what? Maybe getting wasted every once in a while was not the worst thing in the world.\u201d And whether wasted or not, maybe just having a couple of drinks every Friday or Saturday when you\u2019re out and about was part of what helped society get to where it is. Maybe the fact that humans have literally been drinking for whatever it is, like 13,000 years fermenting things, had a social purpose, and we pull out Chesterton\u2019s fence of our own chagrin, right?\n \n \n Lex Fridman \n (05:00:08) \n As you mentioned, in the spirit of that, so when I traveled across rural China, I did smoke a little bit in, in my early 20s, but I picked up just for that trip-\n \n \n DHH \n (05:00:19) \n A social connector, right?\n \n \n Lex Fridman \n (05:00:22) \n And I drank humongous amounts, even though I don\u2019t really drink these days, because of the social thing. It\u2019s the way they show love. And I also ate shitload of carbs, which- \u2026 I\u2019m usually low carb. I just like meat. So just because it\u2019s the way they show love. And actually even, it\u2019s in that society cigarettes, but in most societies, liquor is a thing to do together. And it\u2019s a bond. The benefits that that entails can, in many situations, outweigh the negative consequences like long-term for whatever, however you measure them.\n \n \n DHH \n (05:01:02) \n And this is the myopia of modernity, that we reduce things down to like, well, one glass on this long run study meant that you had a 0.1% greater risk of some heart defect. Yes, but if you didn\u2019t live at all, what\u2019s the point? And again, it\u2019s also easy to flip over the other side. Like the only way to live is to get drunk every weekend and eat a bunch of crap and smoke 40 cigarettes a day. No, it\u2019s not. Like it doesn\u2019t have to be the either/or. This is one of the reasons I stopped wearing my Oura Ring. So I wore the Oura sleep ring for many years, four years. And one day I was just like, \u201cI don\u2019t need to know that I had-\u201d \u2026 \u201ca bad night of sleep.\u201d Like I know. This added, I don\u2019t wanna call it an anxiety, I don\u2019t have a lot of anxiety, but this added just like reminder-\n \n \n DHH \n (05:01:55) \n \u2026 that I had a bad night of sleep, what purpose is it actually playing here? Why do I need to have all these stats? Why do I need to optimize everything? To the point where, again, after that single anecdote, you know, from Diary of a CEO, I was like, \u201cFuck, I kind of want the opposite.\u201d Like- I don\u2019t barely drink at all. I, a glass of wine once a month maybe. Like, do you know what? I should start just on the weekends, get a glass. Maybe two.\n \n \n DHH \n (05:02:25) \n And also, I mean, I\u2019ve never been the person to fret about my diet. This is one of the great privileges of being married to\u2014and I\u2019ll do a quick aside here. Don\u2019t ever call a woman in her 40s a wonderful woman. My wife was so pissed\u2026 that not only last year did I come on her birthday, that was bad enough, but that I called her a wonderful woman on the show, because that sounded like her name was Nancy and she was 65\u2014 \u2026 in the suburb of Rochester. So I\u2019m like, I was trying to play the 50. Like, no, I don\u2019t know, I\u2019m Danish. How, how would I know? But that didn\u2019t go over well. So I\u2019m gonna say awesome woman instead.\n \n \n DHH \n (05:03:07) \n If you marry an awesome woman who also not only knows how to cook, but enjoys it do you know what? There\u2019s a very traditional division of labor here that has been quite stable through several \u2026 tens of thousands of years in human societies, that perhaps also wasn\u2019t the worst idea in the world. And I also know there are plenty of men who like to cook, and there are women who don\u2019t. But as stereotypes go, eh, that certainly helped me. Like, I don\u2019t think I would\u2019ve been able to just be as relaxed about food if I didn\u2019t eat relatively healthy on a regular basis. But the overall point being just getting out of that optimization game without turning into the opposite. You don\u2019t have to be a fat slob on the couch either. You can move around.\n \n \n Lex Fridman \n (05:03:59) \n Yeah. And then the bigger point we\u2019re talking about is the finiteness of life.\n \n \n \n \n Lex Fridman \n (05:04:05) \n And I agree with you on this.\n \n \n DHH \n (05:04:06) \n This was actually a feature, I built it into Omarchy. So memento mori. Remember death. If you pull down the calendar in Amaci, you open it by clicking the clock. It opens up, and the first funny thing, I took it off the internet. There\u2019s this account called Years Progress, and it, every day it updates, and it just shows 63% done with 2026. I don\u2019t know if you\u2019ve seen that one. So it just fills up. So I built that into the calendar. So right now it says, \u201cWe\u2019re 63% done with 2026.\u201d But then if you double-click that, it\u2019ll ask you, \u201cWhen were you born? How long do you expect to live?\u201d And the default is 90. So I put in 1979. And then right below the line of, like, how long are we done with the year-\n \n \n Lex Fridman \n (05:04:47) \n Oh, that\u2019s beautiful.\n \n \n DHH \n (05:04:47) \n \u2026 is a new line that shows how long are you done with your life. I think I\u2019m at 62%. I\u2019m like, if you hover over it, it says, \u201cMemento mori.\u201d\n \n \n \n \n DHH \n (05:04:57) \n Fair number of people found it morbid, but this was why it is an Easter egg. It\u2019s not, like, exposed. You gotta click a few things to get there. I like the reminder. I like the reminder that time is finite, and I should make the most of it. And this is also one of the privileges of having children. You really realize that the cliches are all true. Oh my God, it goes so fast. And I remember my now 13-year-old being one, being two. I was like, \u201cThis was just yesterday.\u201d And the reason it\u2019s a cliche is because it\u2019s a shared experience. Cliches are good. They remind you that you are part of the human experience, and others have been here before you.\n \n \n Eternal recurrence and future of human civization\n \n Lex Fridman \n (05:05:38) \n I have to ask, you mentioned the amazing woman and great cook.\n \n \n \n \n \n \n DHH \n (05:05:44) \n Very important to remember. Not wonderful, amazing.\n \n \n Lex Fridman \n (05:05:47) \n Amazing. So there\u2019s this\u2026 You talked some shit about croissants or something. I saw this in a tweet. So is the Danish croissant somehow better?\n \n \n DHH \n (05:05:59) \n Actually, Danish croissants are not particularly good. In my opinion, the only place you can find a half-decent croissant, of all places, is 7-Eleven, which is hilarious to me because 7-Eleven in the US is the place where, at least in Chicago, you\u2019d go to get a slushie and try to avoid getting murdered. But in Denmark, 7-Eleven is this very upscale, super nice convenience store, and they just happen to make decent croissants. And it always baffled me that I lived in the United States for 20 freaking years. I have yet to have a single croissant that\u2019s even at the level of a 7-Eleven croissant in Copenhagen. That seems improbable.\n \n \n Lex Fridman \n (05:06:41) \n Have you understood why?\n \n \n DHH \n (05:06:43) \n No, I\u2019ve legitimately considered hiring an investigative journalist to get to the bottom of this, because the number of theories I\u2019ve heard have been all over the place. It\u2019s the water. It\u2019s the butter. It\u2019s the skill. It\u2019s the this, it\u2019s the that, it\u2019s the other thing. I\u2019m like, none of it adds up to me. If it\u2019s the butter, can\u2019t you just import it? I thought I saw Danish butter in some supermarkets over here at the Lurpak. So that can\u2019t be it. Can it be the skill? No, it can\u2019t be the skill. There are plenty of French people and Belgian people, the best croissant makers in the world, who live in the US. Like, what is the reason? And then I get the second thing, which is- \u2026 \u201cNo, no, you just haven\u2019t gone to the right place.\u201d I\u2019ve literally gone to 15 of quote unquote-\n \n \n DHH \n (05:07:25) \n \u2026 \u201cthe right place,\u201d and all the croissants were shit. So before I leave the planet, I have to get to the bottom of why the United States cannot make a proper croissant. And I don\u2019t, I\u2019m not even asking about peak French croissants, peak Belgian croissants- \u2026 which are the best croissants in the world. I\u2019m just asking about the 7-Eleven croissant from Copenhagen. Is that too much to ask for?\n \n \n Lex Fridman \n (05:07:49) \n That\u2019s a fascinating mystery. What, zooming out, what\u2019s your favorite\u2026 This makes me wanna know, what\u2019s your favorite meal? Like, if you, if I gave you a last meal and were to murder you after.\n \n \n DHH \n (05:08:02) \n It all depends whether it\u2019s lunch or it\u2019s dinner. If it\u2019s lunch-\n \n \n Lex Fridman \n (05:08:05) \n It\u2019s one meal. What\u2019s\u2026 I feel like if you\u2019re getting murdered, it doesn\u2019t really matter if it\u2019s lunch or dinner.\n \n \n DHH \n (05:08:11) \n Well, it depends on what time of day you\u2019re getting murdered.\n \n \n \n \n DHH \n (05:08:14) \n If it\u2019s an afternoon execution, I only get lunch.\n \n \n \n \n DHH \n (05:08:19) \n And if I get lunch, it\u2019s at Louisiana, the museum north of Copenhagen. I was just there with Jamie two days ago, and we go there all the time. The museum is lovely. We don\u2019t go for the museum. We go for the cafeteria. That in itself sounds crazy. Like-\n \n \n \n \n DHH \n (05:08:40) \n \u2026 you\u2019re gonna, you\u2019re gonna drive half an hour to go to a cafeteria. Well, it just so happens to be, it is the best lunch I\u2019ve probably ever had in my life. The chef there is just unbelievably good, and not only is the food amazing, what\u2019s even more amazing to my systems-thinking brain is that at any one time when we were up there, I think there were probably 200 people waiting for their food.\n \n \n \n \n DHH \n (05:09:04) \n And our food was delivered in five minutes. They have a small menu, and clearly they make it in bulk, and it\u2019s unbelievably delicious.\n \n \n Lex Fridman \n (05:09:15) \n Just a cafeteria at a museum?\n \n \n DHH \n (05:09:17) \n It\u2019s a cafeteria at a museum that happens to have the best lunch I\u2019ve ever had anywhere in my life.\n \n \n Lex Fridman \n (05:09:21) \n What ki- what kind of food are we talking about? What\u2019s, what kind of lunch food?\n \n \n DHH \n (05:09:23) \n It\u2019s, like, Nordic cuisine, so it\u2019s, it\u2019s\u2026 A funny thing is the menu, when I looked at it, I was like, \u201cDo you know what? If I saw this menu\u2026\u201d As a kid, I\u2019d go like, \u201cI\u2019m not eating here.\u201d\n \n \n \n \n \n \n Lex Fridman \n (05:09:33) \n This is pretty fancy.\n \n \n DHH \n (05:09:33) \n \u2026 root free fruit and- \u2026 and onions- \u2026 and all sorts of healthy stuff. That sounds disgusting. And then you sit down and eat it, and the angels sing. But it\u2019s even more particular than that. It\u2019s the butter. The whipped butter that they have there, the best butter anywhere in the world.\n \n \n Lex Fridman \n (05:09:50) \n I think that has to be connected to the croissant mystery too.\n \n \n DHH \n (05:09:52) \n There, there must be some connection. But literally- \u2026 even in Paris, I\u2019ve never had whipped butter that tastes as delicious as the butter does at Louisiana in Copenhagen. Literally, I would drive 30 minutes- \u2026 to have bread and butter at the cafeteria in Louisiana.\n \n \n Lex Fridman \n (05:10:10) \n Well, how\u2019d you discover that? Just one day you went to the museum and-\n \n \n DHH \n (05:10:13) \n I think we just went to the museum-\n \n \n Lex Fridman \n (05:10:15) \n And you tried a whole dish?\n \n \n DHH \n (05:10:15) \n \u2026 and we had the lunch, and we were like\u2026\n \n \n \n \n DHH \n (05:10:20) \n This is rivaling the best tasting restaurants- \u2026 I\u2019ve ever been to in my life, and I\u2019ve gone to Noma a couple of times and tried what there is to try, and this is right up there.\n \n \n Lex Fridman \n (05:10:31) \n All right. What about dinner, since you mentioned? Is it something different that you have to think-\n \n \n DHH \n (05:10:36) \n Well, I think there\u2019s a lot of good options in Copenhagen, but actually I\u2019ll pick a place in Spain. So there is this place in Marbella called Puente Romano. It\u2019s a hotel, and down by the beach, they have this beach restaurant with fresh seafood. Unbelievably delicious. I\u2019m trying to remember what it\u2019s called, but yeah, if you look it up, the beach seafood place at Puente Romano is incredible.\n \n \n Lex Fridman \n (05:11:08) \n How much of great food is, like, the place and the person you\u2019re with, right? Or the experience when you first are there?\n \n \n DHH \n (05:11:18) \n I mean, the ambiance certainly helps, but I think my taste buds are also just-\n \n \n \n \n \n \n \n \n \n \n Lex Fridman \n (05:11:24) \n \u2026 it is just good food.\n \n \n DHH \n (05:11:24) \n It\u2019s just really good food. Because for ex- that cafeteria, there\u2019s a lot of people. Like I wouldn\u2019t rate it as like, oh my God, this is just a place I want to sit for two hours. But the food is so delicious.\n \n \n Lex Fridman \n (05:11:37) \n That\u2019s one of the things I\u2019m, in traveling across the country, I\u2019m realizing to sample without bias, just try things. \u2018Cause you I guess you might just be surprised. Maybe I\u2019ll find your croissant. All right, we\u2019ve talked about mortality. We talked about food.\n \n \n \n \n Lex Fridman \n (05:11:58) \n A thousand years from now, what do you think is the future of humans, human civilization? You think we\u2019re gonna make it?\n \n \n DHH \n (05:12:04) \n A thousand years from now?\n \n \n \n \n DHH \n (05:12:06) \n Man, I\u2019m having trouble predicting 12 months from now.\n \n \n \n \n \n \n Lex Fridman \n (05:12:10) \n You think we got a shot?\n \n \n DHH \n (05:12:11) \n Oh, I\u2019m gonna bet on optimism. I\u2019m gonna bet on a multi-planetary species. I\u2019m gonna bet that Elon gets that rocket to Mars, and we figure out how to bend space-time and discover some wormholes or, or something.\n \n \n Lex Fridman \n (05:12:24) \n Once we do, do you think, you think all humans on Earth will die once or twice and be repopulated after we get a good backup going?\n \n \n DHH \n (05:12:35) \n I mean, that was actually one of my favorite episodes of Black Mirror, the one where they get uploaded to the 1980s sea town somewhere. And I just thought like, wow, this is both such a dark moment, yet also so beautiful in the shared recognition that, like, the \u201980s were actually kind of amazing. Like, I don\u2019t have a lot of nostalgia for many moments, but I do have nostalgia for the \u201980s.\n \n \n \n \n DHH \n (05:13:01) \n And they say something about, like, whatever place and time you were in, the music you were listening to when you were 12 or- \u2026 or 14, 15, something like that, that\u2019s what\u2019s gonna stick with you. Like, no, I was only, I got to be 10 in the \u201980s. So this was my early childhood, and I still have an unbelievably fond affiliation with the \u201980s. So if I\u2019m gonna be uploaded to the sky-\n \n \n Lex Fridman \n (05:13:23) \n You want it to be the \u201980s.\n \n \n \n \n Lex Fridman \n (05:13:26) \n Wow. I don\u2019t think I\u2019ve actually ever heard anyone say that. Usually it\u2019s \u201970s or \u201990s.\n \n \n \n \n Lex Fridman \n (05:13:31) \n Because the \u201980s is a-\n \n \n DHH \n (05:13:32) \n I really don\u2019t like the \u201990s. Can we talk about this?\n \n \n Lex Fridman \n (05:13:38) \n No, I\u2019m done. Not done.\n \n \n DHH \n (05:13:40) \n To me, this was the great turning point of nihilism. That both the music, the genre, the fashion- \u2026 everything turned from, like, this glamour, the pop, the optimism, even the yuppies and everything- \u2026 to grunge and Nirvana. And even though, I mean, I like the music, I don\u2019t like the ethics. I don\u2019t like- \u2026 the morals. I don\u2019t like any of the underpinnings to it. There was just such a defeatism to it- \u2026 that in retrospect, I look back upon it, if I\u2019m gonna get reinserted into the matrix, it\u2019s gonna be the \u201980s.\n \n \n Lex Fridman \n (05:14:17) \n That was full of optimism and fun- \u2026 and colors.\n \n \n DHH \n (05:14:22) \n Yes. I remember having these orange pants- \u2026 with, like, white dots on them. That was just a normal thing kids could wear in the \u201980s. I\u2019ve never fucking seen that today. Or I mean, even in the \u201990s, it all turned just, like, Seattle gray. Great regression.\n \n \n Lex Fridman \n (05:14:46) \n So the Nietzschean, you know, this idea of eternal recurrence. If you do a Groundhog Day and you live forever, you want it to be in the orange pants with the white dots in the \u201980s.\n \n \n DHH \n (05:14:56) \n Maybe I\u2019m gonna skip the orange pants, but- \u2026 I\u2019m definitely picking the \u201980s.\n \n \n Lex Fridman \n (05:14:58) \n You said it. All right, DHH. Thank you so much for everything you do. Thank you for being an inspiration to all of us, for building cool shit in the world, and thank you for talking today.\n \n \n DHH \n (05:15:11) \n Anytime. Thank you, Lex, again for having me.\n \n \n Lex Fridman \n (05:15:15) \n Yeah. Thanks for listening to this conversation with DHH. To support this podcast, please check out our sponsors in the description, where you can also find links to contact me, ask questions, give feedback, and so on. And now, let me leave you with some words from Ralph Waldo Emerson. \u201cOnce you make a decision, the universe conspires to make it happen.\u201d Thank you for listening, and hope to see you next time." ], "output": "readability", "pwd": "/data/archive/1788689050.032601", "schema": "ArchiveResult", "start_ts": "2026-09-06T10:06:47.990450+00:00", "status": "succeeded" } ], "screenshot": [], "singlefile": [], "title": [ { "cmd": [ "curl", "--silent", "--location", "--compressed", "--max-time", "120", "--user-agent", "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) curl/curl 8.14.1 (x86_64-pc-linux-gnu)", "https://lexfridman.com/dhh-2-transcript/" ], "cmd_version": "curl 8.14.1 (x86_64-pc-linux-gnu)", "end_ts": "2026-09-06T10:06:47.982109+00:00", "index_texts": null, "output": "Transcript for DHH: Future of Programming, AI, Agentic Engineering, Vibe Coding & Linux | Lex Fridman Podcast #501 - Lex Fridman", "pwd": "/data/archive/1788689050.032601", "schema": "ArchiveResult", "start_ts": "2026-09-06T10:06:47.800199+00:00", "status": "succeeded" } ], "wget": [ { "cmd": [ "wget", "--no-verbose", "--adjust-extension", "--convert-links", "--force-directories", "--backup-converted", "--span-hosts", "--no-parent", "-e", "robots=off", "--timeout=120", "--restrict-file-names=windows", "--warc-file=/data/archive/1788689050.032601/warc/1788689196", "--page-requisites", "--user-agent=Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36 ArchiveBox/0.7.4 (+https://github.com/ArchiveBox/ArchiveBox/) wget/GNU Wget 1.21.3", "--compression=auto", "https://lexfridman.com/dhh-2-transcript/" ], "cmd_version": "GNU Wget 1.21.3", "end_ts": "2026-09-06T10:06:47.787614+00:00", "index_texts": null, "output": "lexfridman.com/dhh-2-transcript/index.html", "pwd": "/data/archive/1788689050.032601", "schema": "ArchiveResult", "start_ts": "2026-09-06T10:06:36.513857+00:00", "status": "succeeded" } ] }, "icons": null, "is_archived": true, "is_static": false, "latest": { "archive_org": null, "dom": null, "favicon": "favicon.ico", "git": null, "media": null, "pdf": null, "screenshot": null, "singlefile": null, "title": "Transcript for DHH: Future of Programming, AI, Agentic Engineering, Vibe Coding & Linux | Lex Fridman Podcast #501 - Lex Fridman", "warc": null, "wget": "lexfridman.com/dhh-2-transcript/index.html" }, "link_dir": "/data/archive/1788689050.032601", "newest_archive_date": "2026-09-06T10:06:54.939116+00:00", "num_failures": 0, "num_outputs": 6, "oldest_archive_date": "2026-09-06T10:06:35.573123+00:00", "path": "/dhh-2-transcript/", "schema": "Link", "scheme": "https", "snapshot_id": "c017c156-8c67-440f-933a-b102d25b29c5", "sources": [ "/data/sources/1788689050-import.txt" ], "tags": null, "tags_str": "", "timestamp": "1788689050.032601", "title": "Transcript for DHH: Future of Programming, AI, Agentic Engineering, Vibe Coding & Linux | Lex Fridman Podcast #501 - Lex Fridman", "updated": "2026-09-06T10:06:35.572488+00:00", "updated_date": "2026-09-06 10:06", "url": "https://lexfridman.com/dhh-2-transcript/" }