Skip to content
Y CombinatorY Combinator

Boris Cherny: How Tool Use Turned a CLI Into Claude Code

By betting on bash tools and a minimal terminal loop; CLAUDE.md emerged from organic user demand, and subagents now parallelize repo exploration.

Boris ChernyguestGarry TanhostJared Friedmanhost
Feb 17, 202650mWatch on YouTube ↗

EVERY SPOKEN WORD

  1. 0:00 – 1:45

    Intro

    1. BC

      at Anthropic, the way that we thought about it is, we don't build for the model of today, we build for the model six months from now. That's actually, like, still my advice to, to founders that are building on LLMs. Just try to think about, like, what is that frontier where the model is not very good at today? 'Cause it's gonna get good at it. All of Claude Code has just been written and rewritten and rewritten and rewritten over and over and over. There is no part of Claude Code that was around six months ago. You try a thing, you give it to users, you talk to users, you learn, and then eventually you might end up at a good idea. Sometimes you don't.

    2. GT

      Are you also in the back of your mind thinking that maybe, like, in six months you won't need to prompt that explicitly, but the model will just be good enough to figure out on its own?

    3. BC

      Maybe in a month. [laughing]

    4. JF

      [laughing] No more need for plan mode in a month?

    5. BC

      [laughing]

    6. GT

      Oh, my God.

    7. JF

      [music]

    8. GT

      Welcome to another episode of The Lightcone, and today we have an extremely special guest, Boris Cherny, the creator, engineer of Claude Code. Boris, thanks for joining us.

    9. BC

      Thanks for having me.

    10. GT

      Thanks for creating a thing that has taken away my sleep for about three weeks straight. [laughing]

    11. JF

      [laughing]

    12. BC

      [laughing]

    13. GT

      I am very addicted to Claude Code, and, uh, it feels like rocket boosters. Has it felt like this for people, like, for, you know, months at this point? I think it was, like, end of November is where, uh, a lot of my friends said, like, something changed.

    14. BC

      I remember for me, I felt this way when I first created Claude Code, and I didn't yet know if I was onto something. I kinda felt like I was onto something, and then that's when I wasn't sleeping. [laughing]

    15. GT

      Yeah.

    16. BC

      And that was just like-

    17. GT

      When was that?

    18. BC

      - three straight months. This was, uh, September twenty twenty-four. Yeah, it was, like, three straight months. I, I didn't take a single day of vacation, worked through the weekends, worked every single night. I was just like, "Oh, my God, this is-- I think this is gonna be a thing. I don't know if it's useful yet," [chuckles] 'cause it, it couldn't actually code yet.

  2. 1:45 – 2:38

    The most surprising moment in the rise of Claude Code

    1. GT

      If you look back on, uh, those moments to now, like, what would be, like, the most surprising thing about this moment right now?

    2. BC

      It's unbelievable that we're still using a terminal. That was supposed to be the starting point. I didn't think that would be the ending point. And then the second one is that it's even useful, 'cause, uh, you know, at the beginning, it didn't really write code. Even in February, when we cheated, it wrote maybe, like, ten percent of my code or something like that. I didn't really use it to write code. It wasn't very good at it. I still wrote most of my code by hand. Uh, so the fact that it, it-- actually, like, our bets paid off, and it got good at the thing that we thought it was gonna get good at, because it wasn't obvious. At Anthropic, the way that we thought about it is, we don't build for the model of today. We build for the model six months from now, and that's actually, like, still my advice to, to founders that are building on LLMs is, you know, just try to think about, like, what is that frontier where the model is not very good at today? Um, 'cause it's gonna get good at it, and you just have to

  3. 2:38 – 5:38

    How Boris came up with the idea for Claude Code

    1. BC

      wait.

    2. GT

      Going back, but when-- do you remember when you first got the idea? Can you just talk us through that? Like, was there some, like, a spark, or what was even the first version of it in your mind?

    3. BC

      You know, it's funny. It was like, it was so accidental that it, it just kinda evolved into this. Um, you know, a, a, as Anthropic, I think for Anth, the bet has been coding for a long time, and the bet has been the path to save, to safe AGI is through coding.

    4. GT

      Mm-hmm.

    5. BC

      And this is, this has kinda always been the idea. And the way you get there is you, you teach the model how to code, then you teach it how to use tools, then you teach it how to use computers. Um, and you can kinda see that, 'cause the, the first team that I joined at Anthropic, it was called Anthropic Labs team, uh, and it produced three products. It was Claude Code, MCP, and the desktop app. So you can kinda see how these like weave together. The particular product that we built, you know, like, no one, no one asked me to build a CLI. [chuckles] Um, we kind of knew maybe it was time to build some kinda coding product, 'cause it seemed like the model was ready, but no one had yet really built the product that harnessed this capability. So, like, still there's this insane feeling of product overhang, but at the time, it was just, like, even crazier 'cause, like, no one had built this yet. And so I, I started, like, hacking around, uh, and the, the... I was like, "Okay, we build a coding product. What do I have to do first? I have to understand how to use the API," 'cause I hadn't used the Anthropic API at that point. Um, and so I, I just built, like, a little terminal app to use the API. That, that's all that it did, and it was a little chat app, 'cause, you know, like, you think about the, you know, AI applications at the time, and, you know, for non-coders today, most... What are, what are most people using? It's just a chat app, so that's what I built. Uh, and, you know, it was in a terminal. I can ask questions. I get the answers. Then I think tool use came out. I just wanna try out tool use, 'cause I, I don't really understand what this is. I was like, "Tool use, this is cool. Is this actually useful? Probably not. Let me just try it."

    6. GT

      You built it in terminal just because it was the easiest way to get something up and running?

    7. BC

      Yes, 'cause I didn't have to build a UI.

    8. GT

      Okay.

    9. BC

      So it was just me.

    10. GT

      At that point, it was, like, the IDEs, Cursor, Windsurf, were the things that were really taking off. Were you sort of under any pressure or getting lots of suggestions of, "Hey, like, we should build this out as a plug-in or as a, as a fully featured IDE itself?"

    11. BC

      There was no pressure 'cause we didn't even know what we wanted to build. Like, the, the team was just in explore mode, you know? Like, we, we didn't-- we know vaguely we wanted to do something in coding, but it wasn't obvious what. No one was high confidence enough. That was, like, my job to figure out. And so I gave, I gave the model, uh, the bash tool. That was the first tool that, that I gave it, just 'cause I think that was literally the example in our docs. [laughing]

    12. GT

      [laughing]

    13. BC

      I just, like, took the example. It was in Python. I just ported it to TypeScript, 'cause that, that's how I wrote it. You know, I didn't know, like, what the model could do with bash, so I asked it to, like, read a file. It could, like, cat the file, so, like, that was cool. And then I was like: Okay, like, what can you actually do? And I, and I asked it, "What music am I listening to?" It wrote some, like, AppleScript to script my, my Mac and look up the music in my music player.

    14. GT

      Oh, my God.

    15. BC

      And this was Sauna three point five.

    16. GT

      Mm.

    17. BC

      And, you know, like, I, I didn't think the model could do that. And that was my first, I think, ever fuel the AGI moment-

    18. GT

      Mm-hmm

    19. BC

      ... where it's just like: Oh, my God, the model, it, it just wants to use tools. That, that's all it wants.

  4. 5:38 – 7:09

    The elegant simplicity of terminals

    1. JF

      That's kind of fascinating. I mean, it's very kinda contrarian that Claude Code works so well in such an elegant, simple form factor. I mean, terminals have been around for a really long time, and that seemed to be like a good design constraint that allowed a lot of interesting developer experiences. Like, you, y- it doesn't feel like working. It just feels fun. As a developer, I don't have to think about files, where everything is.... and that came by accident almost?

    2. BC

      Yeah, it was an accident. I remember- so after the terminal started to take off internally, um, and, uh, honestly, like, after building this thing, I think like two days after the first prototype, I started giving it to my team just for dogfooding. 'Cause, uh, you know, like, you know, i- if you come up with an idea and it seems useful, the first thing you wanna do is you wanna give it to people to see how they use it. And then I came in the next day, and then Robert, who sits across from me, who's another engineer, he w- he just, like, had Claude Code on his computer, and he was, like, using it to code. [laughing]

    3. GT

      [laughing]

    4. BC

      I was like- I was like: "What? What are you- what are you doing? Like, this thing isn't ready. It's just a prototype." But yeah, it w- it was already useful in that form factor. And I remember when we did our launch review to kinda launch Claude Code externally, this was in December, Nov- November, or something like that, in twenty twenty-four. Um, Dario asked, and he was like: "The usage chart internally, like, the, the Dow chart is, like, vertical. Are you, like, forcing engineers to use it?" [chuckles] "Like, why are you mandating them?" And I was just like: "No, no, we didn't." We d-- I just, like, posted about it, and they, they'd just been, like, telling each other about it. Honestly, it was, it was just accidental. We, we started with the CLI 'cause it was the cheapest thing, and it just kinda stayed there

  5. 7:09 – 9:00

    The first use cases

    1. BC

      for a bit.

    2. GT

      So in that twenty twenty-four period, w- what- how were the engineers using it? Were they sort of shipping code with it yet, or were they using it in a different way?

    3. BC

      The model was not very good at coding yet. I w- I was using it personally for automating Git.

    4. GT

      [laughing]

    5. JF

      [laughing]

    6. BC

      Um, I think at this point, I, I've probably forgotten most of my Git because- [laughing]

    7. GT

      [laughing]

    8. BC

      - Claude Code has just been doing it for so long. But yeah, like automating, uh, bash commands, that w- that was a very early use case, and, like, operating like Kubernetes and kinda things like this. People were using it for coding, so there were some early signs of this. I think the first use case was actually writing unit tests-

    9. GT

      Hmm

    10. BC

      ... 'cause it's a little bit lower risk, and the model was still pretty bad at it. [laughing] But people were, were, were kinda figuring it out, and they- and they were figuring out how to use this thing. Um, and one thing that we saw is people started writing these markdown files for themselves and then having the model read that markdown file, and this is where Claude MD came from. Probably, the single, for me, biggest principle and product is latent demand. Um, and it- just every bit of this product is built through latent demand after the initial CLI, uh, and so Claude MD is an example of that. There's this other general principle that I think is maybe interesting, where you can build for the model, and then you can build scaffolding around the model in order to improve performance a little bit. And depending on the domain, you can improve performance maybe ten, twenty percent, something like that, and then essentially the gain is wiped out with the next model. So either you can build the- build the scaffolding and then, you know, get some performance gain and then rebuild it again, or you just wait for the next model, and then you kinda get it for free. The Claude MD and kinda the scaffolding is an example of that, and really, I think that's why we stayed in the CLI, is because we felt there is no UI we could build that would still be relevant in six months-

    11. GT

      Hmm

    12. BC

      ... because the model was improving so quickly.

    13. GT

      Uh, earlier, we were saying, like, we should compare Claude MDs, but you said something very profound, which is, you know, yours is actually very short, which is almost, like, the opposite of what, you know, people might expect. Why

  6. 9:00 – 11:29

    What’s in Boris’ CLAUDE.md?

    1. GT

      is that? What's in your Claude MD?

    2. BC

      Okay, so I, I checked this before we came. So my, my Claude MD has two things. Um, one is, uh, there- it, it, it's just two lines. So the first line is, "Whenever you put up a PR, enable auto-merge," um, so as soon as someone accepts it, it's merged. That's just so I can, like, code, and I don't have to kinda go back and forth with CR or whatever. And then the second one is, "Whenever I put up a PR, post it in our internal team stamps channel," uh, just so someone can stamp it, and I can get unblocked. Uh, and the idea is every other instruction is in our Claude MD that's checked into the code base, and it's something our entire team contributes to multiple times a week. And very often, I'll see someone's PR, and they, they make some, like, mistake that's totally preventable, and I'll just literally tag Claude on the PR. I'll just do, like, add Claude, you know, like, add this to the Claude MD, and I'll do this, you know, like, many times a week.

    3. GT

      Do you have to, like, compact the Claude MD? Like, I've definitely reached a point where I got the message at the top saying: "Your Claude MD is, like, thousands of tokens now." What do you do when you guys hit that?

    4. BC

      So our Claude MD is actually pretty short. I think it's, like, couple thousand tokens, maybe something like that. Um, if you, if you hit this, my recommendation would be delete your Claude MD and just start fresh.

    5. GT

      Interesting.

    6. JF

      Hmm.

    7. BC

      I think a lot of people, like, they try to over-engineer this, right? And, and really, like, the capability changes with every model, and so the thing that you want is do the minimal possible thing in order to get the model on track. And so if you delete your Claude MD, and then, you know, the model is getting off track, it does the wrong thing, that's when you kinda add back a little bit at a time. And you- what you're probably gonna find is, with every model, you have to add less and less. For me, I consider myself a pretty average engineer, to be honest. Like, I don't use a lot of fancy tools. Like, I, I don't use, like, Vim. I use, you know, VS Code because it's simpler. Um, I don't really-

    8. GT

      Wait, wait, wait, really? I would have assumed that because you built this in the terminal, that you were sort of like a die-hard ter- terminal, like Vim, Vim-only person, you know, screw those VS Code, Code people, you know. [chuckles]

    9. BC

      Well, we have people like that on the team. There's, you know, like Adam Wolf, for example, he's on- he's on the team. He's like: "You will never take Vim from my cold, dead hands."

    10. GT

      [laughing]

    11. JF

      [laughing]

    12. BC

      Like... Yeah, so there's definitely a lot of people like that on the team, and this is one of the things that I learned early on, is every engineer likes to hold their dev tools differently. They like to use different tools. There's just no one tool that works for everyone. But I think also, this is one of the things that makes it possible for Claude Code to be so good because I kind of think about it as, what is the product that I would use that makes sense to me? And so to use Claude Code, you don't have to understand them. You don't have to understand Tmux. You don't have to know how to, like, SSH. You don't have to know all the stuff. You just have to open up the tool, and it'll guide you. It'll, it'll do all the stuff.

    13. GT

      How

  7. 11:29 – 15:44

    How do you decide the terminal’s verbosity?

    1. GT

      do you decide how verbose you want, like, sort of the terminal to be? Like, sometimes you have to go, c- you know, Control-O, and check it out, and is it like internal bike shed battles around, [chuckles] like, longer or shorter? I mean, every, every user probably has a different opinion. Like, how do you make those sorts of decisions?

    2. BC

      What, what's your opinion? Is it, is it too verbose right now?

    3. GT

      Oh, I love the verbosity 'cause basically, sometimes it just, like, goes off the deep end, and I'm watching, and then I can just read very quickly, and it's like, "Oh, no, no, it's not that," and then I escape and then just stop it, and then it just, like, stops an entire bug farm, like, as it's happening. I mean, that's usually when I didn't do plan mode properly. [chuckles]

    4. BC

      This is something that we probably change pretty often.... um, I remember early on, this is maybe six months ago, I tried to get rid of bash output just internally, just to, like, summarize it, 'cause I was like, "Eh, these giant long bash commands, I don't actually care." And then I gave it to Anthropic employees for a day, and everyone just revolted. [laughing]

    5. GT

      [laughing]

    6. BC

      They were like, "I wanna see my dash." 'Cause it, it actually is quite useful for, you know, like, uh, for something like git output, maybe it's not useful, but if you're running, you know, like, Kubernetes jobs or something like this, you actually do wanna see it. We recently hid the, hid the file reads and, uh, file searches, so you'll notice instead of saying, you know, like, "Read foo.md," it'll said, you know, like, "Read one file, search, searched one pattern." And this is something I think we could not have shipped six months ago, 'cause the model just was not ready. It would've, you know, it still read the wrong thing pretty often. As a user, you still had to be there and kinda catch it and debug it. But nowadays, I just noticed it's on the right track almost every time. And because it's using tools so much, it's actually a lot better just to summarize it. Um, but then we shipped it, uh, we dogfooded it for, like, a month, and then people on GitHub didn't like it. Uh, so there was a big issue where people were like, "No, like, I wanna see the details," and that was really great feedback. Um, and so we added a new verbose mode, and so that's just, like, in slash config, you can enable verbose mode, and if you wanna see all the file outputs, you can continue to do that. And then I posted on the issue, and people still, still didn't like it, which is, again, awesome, because, like, my favorite thing in the world is just hearing people's feedback and hearing how they actually wanna use it. Um, and so we just, like, iterated more and more and more to get that really good and to make it the thing that people want.

    7. GT

      I'm amazed, like, how much I enjoy b- uh, fixing bugs now, and then all you have to do is, uh, have really good logging, and then even just say, like, "Hey, check out that, you know, this particular object, it messed up in this way." [chuckles] And it, like, searches the log. It figures everything out. It can, like, go into your... You can make a production tunnel, and it'll look at your production DB for you. It's like, this is insane. Bug fixing is just going to Sentry, copy markdown, you know? Pretty soon-

    8. BC

      [chuckles]

    9. GT

      ... it's just gonna be straight MCP. It's like an auto bug-fixing, like, and test-making sort of, uh... W- what's the new, uh, term they call it? Like a making a startup factory.

    10. BC

      Oh, yeah.

    11. GT

      Right? There's, like, all these concepts now of rather than having to review the code, you know... I'm, I'm old school, so I like the verbosity. I like to say, "Oh, well, you're doing this, but I want you to do that," right? But there's a totally different school of thought now that says, like, anytime an eng- a real human being has to look at code, uh, that's bad. [chuckles]

    12. BC

      Yeah, yeah, yeah, and I-

    13. GT

      Which is fascinating.

    14. BC

      I think, like, Dan Chipper talks about this a lot as kind of when- whenever you see the model make a mistake, try to put it in the CLAUDE.md, try to put it in, like, skills or something like that, so it's reusable. But I, I think there's this meta point that I actually struggle with a lot, and I- people talk about, like, agents can do this, agents can do that, but actually, what agents can do, it changes with every single model. And so sometimes there's a new person that joins the team, and they actually use Claude Code more than I would've used it.

    15. GT

      Hmm.

    16. BC

      And I'm just constantly surprised by this. Like, for example, there was a... We had, like, a memory leak, and we were trying to debug it, um, and by the way, like, Jared Sumner has just been on this crusade, killing all the memory leaks, and it's just been amazing. But before Jared was on the team, I had to do this, and there was this memory leak. I w- I was trying to debug it, and so I t- I took a heap dump. I opened it in dev tools. I was looking through the profile, then I was looking through the code, and I w- I was just trying to figure this out. And then another engineer on the team, Chris, he just, like, asked Claude Code. He was like: "Hey, I think there's a memory leak. Can you, like, run this-

    17. GT

      [chuckles]

    18. BC

      ... and then, like, try to figure it out?" And Claude Code, like, took the heap dump. It wrote a little tool for itself to, like, analyze the heap dump, [chuckles] and then it found the leak faster than I did. And this is just something I have to constantly relearn because my brain is still stuck somewhere six months ago at times.

  8. 15:44 – 18:56

    Beginner’s mindset is key as the models improve

    1. JF

      So what would be some advice for technical founders to really become maximalists at the latest model release? It sounds like people off of, fresh off of school or that, that don't have any assumptions might be better suited than maybe sometimes engineers who have been working at it for a long time. And how do the experts get better?

    2. BC

      I think for yourself, it's kinda beginner mindset, and, uh, I don't know, maybe just, like, humility. Like, I, I feel like engineers, as a discipline, we've learned to have very strong opinions, and senior engineers are kind of rewarded for this. In my old job at a big company, when I hired, like, architects and this kind of a type of engineer, you look for people that have a lot of experience and really strong opinions. But it actually turns out a lot of this stuff just isn't relevant anymore, and a lot of these opinions should change because the model is getting better. Um, so I think, actually, the, the biggest skill is people that can think scientifically and can just think from first principles.

    3. JF

      How do you s- screen for that when you try to hire someone now for, for your team?

    4. BC

      I sometimes ask about: What's an example of when you're wrong? It's a really good one. Uh, you know, some of these, like, classic behavioral questions, like, not even coding questions, I think are quite useful. 'Cause you can see if people can recognize their mistake in hindsight, if they can claim credit for the mistake, and if they learned something from it. And I think a lot of these, like, very senior people, especially, there, there are some founder types like this, but I think founders, in particular, are actually quite good at it. Um, but other people sometimes will never really take, uh... they'll never take the blame for a mistake. But I don't know, like, for me, personally, I'm wrong probably half the time. Like, half my ideas are bad, and you just have to try stuff, and, you know, you try a thing, you give it to users, you talk to users, you learn, and then eventually, you might end up at a good idea. Sometimes you don't.... and this is the skill that I think in, in the past was very important for founders, but now I think it's very important for every engineer.

    5. GT

      Do you think, um, you would ever hire someone based on the, uh, Claude Code transcript of, uh, them working with the agent? 'Cause we're-

    6. BC

      Ask your friends.

    7. GT

      - actively doing that right now. [chuckles]

    8. BC

      Yeah.

    9. GT

      We just added, uh, just as a test, like, you can upload a transcript of you coding a feature with Claude Code or Codex or whatever it is. Personally, I think that, like, it's gonna work.

    10. BC

      Mm.

    11. GT

      I mean, you can figure out, uh, how someone thinks, like, whether they're looking at the logs or not. Like, can they correct the agent if it goes off, off the rails?

    12. BC

      Mm.

    13. GT

      Like, do, do they use plan mode? You know, when they use plan mode, do they make sure that there are tests or are... You know, all of these different things that, you know, do they think about systems? Do they even understand systems?

    14. BC

      Mm.

    15. GT

      Like, there's just so much that's sort of embedded in that, that I imagine. I just want like a spider, uh, a spider web graph, you know, like in those video games, like NBA 2K-

    16. BC

      Mm

    17. GT

      ... it's like, oh, this person's really good at shooting or defense. It's like, you can imagine a spider web graph of like, you know, someone's Claude Code skill level. [chuckles]

    18. BC

      Yeah. What would the, what would the skills be? What would be those axes?

    19. GT

      I mean, I think it's like systems, testing, must be like user behav-- I mean-

    20. BC

      Mm

    21. GT

      ... there's gotta be a design part-

    22. BC

      Yeah

    23. GT

      ... yeah, for sure.

    24. BC

      Like product sense.

    25. GT

      Yeah, product sense.

    26. BC

      Maybe, maybe also just, like, automating stuff.

    27. GT

      Mm-hmm. My favorite thing in Claude MD, uh, for me, is I have a thing that says, "For every plan, decide whether it's over-engineered, under-engineered, or perfectly engineered, and why." [chuckles]

    28. BC

      I think this is something that we're trying to figure out, too, 'cause I, I think, uh,

  9. 18:56 – 21:51

    Hyper specialists vs hyper generalists

    1. BC

      when I look at engineers on the team that I think are the most effective, there's essentially two... It's very bimodal. Um, there's one side where it's extreme specialists, um, and so, like I named Jared before, like, he's a really good example of this, and kinda the Bun team is a really good example. Just hyper specialists, they understand dev tools better than anyone else. They understand JavaScript runtime systems better than anyone else. And then there's the flip side of kinda hyper generalists, and that's kinda the rest of the team. And a lot of people, they span, like, product and infra or product and design, um, or, you know, like product and user research, product and business. I really like to see people that just do weird stuff. I think that's one of these things that was kind of a warning sign [chuckles] in the past, 'cause it's like, can these people actually build something useful?

    2. GT

      Mm-hmm, um, that's the limits test.

    3. BC

      Yeah, that's the limits test. But, but nowadays, they, uh, like, for example, uh, an engineer on the team, Daisy, she was on a different team, and then she transferred onto our team, and the reason that I wanted her to transfer is she put up a PR for Claude Code, like, a couple weeks after she joined or something, and the PR was to add a new feature to Claude Code. And then, instead of just adding the feature, what she did is first she put up a PR to give Claude Code a tool so that it can test an arbitrary tool and verify that that works, and then she put up that PR. And then she had Claude write its own tool instead of herself implementing it.

    4. GT

      Mm.

    5. BC

      And I think it's this kind of out-of-the-box thinking that is, is just so interesting, 'cause not a lot of people get it yet. You know, like, we use the Claude Agent SDK to automate pretty much every part of development. It automates code review, security review, uh, it labels all of our issues, it shepherds things to production. It does pretty much everything for us. But I think externally, I'm seeing a lot of people start to figure this out, but it's actually taken a while to figure out how do you use LMS in this way? How do you use this new kind of automation? So it's kind of a new skill.

    6. GT

      I guess one of the, uh, funnier things that I've been having office hours with various founders about is, um, you, you have, like, sort of the visionary founder who has, like, the idea. They've, like, built this, like, crystal palace of h- the product that they wanna build. They've totally loaded in their brain, you know, who the user is and what they feel and what they're motivated by, and then they're sitting in Claude Code, and they can do, like, you know, fifty X work. And then, but they have engineers who work for them, who, like, don't have the, you know, crystal memory palace of, like, the platonic ideal of the product that the pro- founder has, and they can only do, like, five X work. Are you hearing stories like that? There's usually a person who's, like, the core, like, designer of a thing, and they're just, like, you know, trying to blast it out of their brain. What's the nature of, like, teams like that? You know, it, it seems like that's almost a stable configuration. Like, you're gonna have the visionary, who, like, now is unleashed, but, you know, maybe going back to the top of it, like, I'm experiencing this right now. It's like, oh, well, I'm only a solo person, and, you know, I need to eat and sleep, and I have, you know, a whole job. [chuckles] And it's like, how am I gonna do this, you know?

  10. 21:51 – 23:48

    The vision for Claude teams

    1. BC

      You know, like, we just launched Claude Teams, and, you know, this is a way to do it, but you can also just build your own way to do it. It's pretty easy.

    2. GT

      What's the vision for Claude Teams?

    3. BC

      Just collaboration.

    4. GT

      Yeah.

    5. BC

      It's like, uh, there's this whole new field of, like, agent topologies that people are exploring. Like, what are the ways they can configure agents? There's this one sub-idea, which is uncorrelated context windows, and the idea is just multiple agents. They have fresh context windows that aren't necessarily polluted with each other's context or their own previous context. And if you throw more context at a problem, that's like a form of test time compute. Um, and so y- y- you just get more capability that way, and then, if you have the right topology on top of it, so the agents can communicate in the right way, they're laid out in the right way, then they can just build bigger stuff. And so Teams is kind of like one idea. There's a few more that are coming pretty soon. Um, and the idea is just maybe it can build a little bit more. I think the first kinda big example where it worked is our plugins feature was entirely built by a swarm over, over a weekend.

    6. GT

      Hmm.

    7. BC

      It just ran for, like, a few days. There wasn't really human intervention, and plugins is pretty much in the form that it was when, when it came out.

    8. GT

      How did you set that up? Like, did you spec out sort of the outcome that you were hoping for and then let it sort of figure out the details and then, like, let it run?

    9. BC

      Yeah, an engineer on the team just gave, uh, gave Claude a spec and, um, told Claude to use a Asana board, and then Claude just put up a bunch of tickets on Asana and then spawned a bunch of agents, and the agents started picking up tasks. The main Claude just gave it instructions, and they all just figured it out.

    10. JF

      Like independent, um, agents that didn't have the context of the bigger spec, right?

    11. BC

      Right. If you, if you think about the way that, uh-... you know, like how are agents actually started nowadays? And, you know, I haven't pulled the data on this, but I would bet the majority of agents are actually prompted by Claude today in the form of, uh, subagents. 'Cause, like, a subagent is just like a recursive Claude code. That's all it is in the code, and it's just prompted by, we call her Mama Claude.

    12. GT

      [chuckles]

    13. BC

      Um, and that, that's all it is. And I, I think probably if you look at most agents, they're launched in this way.

    14. GT

      My Claude

  11. 23:48 – 25:12

    Subagents

    1. GT

      Insights just told me to do this more for debugging. So that I get, like, I spend a lot of time on debugging, and it would just be better to have, like, multiple subagents spin up and, like, debug something in parallel. And so then I just, like, added that to my Claude MD to just be like, "Hey, like, next time you're trying to fix a bug, like, have one agent that, like, looks in the log, like, one that looks in the code path." That just seems sort of inevitable. For weird, scary bugs, I try to, uh, fix bugs in plan mode, and then it seems to use the agents to sort of-

    2. BC

      Yep

    3. GT

      ... search everything. Whereas, like, when you're just trying to do it in line, it's like, "Okay, I'm gonna do, like, this one task," instead of search wide.

    4. BC

      This is something I do all the time, too. I j- I just say, if the, if the task seems kinda hard, this kinda research task, I'll calibrate the number of subagents I ask it to use based on the difficulty of the task.

    5. GT

      Hmm.

    6. BC

      So if it's, like, really hard, I'll say, like, "Use three or maybe five or even ten subagents. Research in parallel, and then see what they come up with."

    7. GT

      I'm curious, so then why don't you put that in your Claude MD file?

    8. BC

      It's kinda case by case. You know, like, Claude MD, like, what is it? It's just a... It's a shortcut. Like, if you find yourself repeating the same thing over and over, you put in the Claude MD. But otherwise, you don't have to put everything there. You can just prompt Claude.

    9. GT

      Are you also in the back of your mind thinking that maybe, like, in six months you won't need to prompt that explicitly, like the model will just be good enough to figure out on its own?

    10. BC

      Maybe in a month. [laughing]

    11. JF

      [chuckles] No more need for plan mode in a month?

    12. GT

      [chuckles] Oh, my God!

    13. BC

      I think plan mode probably has a limited lifespan.

    14. GT

      Interesting.

    15. BC

      Oh, yeah.

    16. JF

      There's some alpha for everyone

  12. 25:12 – 28:38

    A world without plan mode?

    1. JF

      here. What would the world look like without plan mode? Do you just describe it at the prompt level, and it would just do it, one-shot it?

    2. BC

      Yeah. We've, uh, we've started experimenting with this 'cause Claude Code can now enter plan mode by itself. I don't know if you've g- you guys have seen that.

    3. GT

      I've seen that, yeah.

    4. JF

      Yeah.

    5. BC

      So we're, we're trying to kind of get this experience really good. So it would enter plan mode the same point where a human would've wanted to enter it. So I think it's like, I think it's something like this, but actually, plan mode, there's no, there's no big secret to it. All it does is it adds one sentence to the prompt that's like, "Please don't code." [chuckles] That's all it is. You can, you can actually just say that.

    6. GT

      Yeah.

    7. JF

      So it sounds like a lot of the feature development for Claude Code is very much, uh, when we talk about a YC, talk to your users-

    8. BC

      Mm.

    9. JF

      -and then you come and implement it. It wasn't the other way, that you had this master plan and then implemented all the features.

    10. BC

      Yeah, yeah. I mean, that, that's all it was. Like, plan mode was, we saw users that, that were like, "Hey, Claude, come up with an idea. Plan this out, but don't write any code yet." And there was kinda various versions of this. Sometimes it was just talking through an idea, sometimes it was these very sophisticated specs that, that they were asking Claude to write. But the common dimension was: do a thing without coding yet. And so literally, like, this was, like, Sunday night at ten pm. I was, I was just, like, looking at GitHub issues and kinda seeing what people were talking about and looking at our internal Slack feedback channel, and I just wrote this thing in, like, thirty minutes and then, uh, shipped it that night. It went out Monday morning, and that was plan mode.

    11. GT

      So do you mean that there will be no need for plan mode to, in the sense of, I'm worried that the model's gonna do, like, it- it's gonna do, like, the wrong thing or head off in the wrong direction, but there will still be a need for that? Y- you need to think through the idea and figure out exactly what it is that you want, and you have to do that somewhere.

    12. BC

      I kinda think about it in terms of, like, kinda increasing model capabilities. So maybe six months ago, a plan was insufficient. So you get Claude to make a plan, let's say even with plan mode, you still have to kinda sit there and babysit 'cause it can go off track. Nowadays, what I do is probably eighty percent of my sessions, I say, I say plan mode has a limited lifespan, but I am, I'm a heavy plan mode user. Um, I- probably eighty must- percent of my sessions, I start in plan mode, and Claude will, you know, it'll start, it'll start making a plan. I'll move on to my second terminal tab, and then I'll have it make another plan, and then when I run out of tabs, I open the desktop app, and then I go to the Code tab, and then I just start a bunch of tabs there. And they all start in plan mode, probably, you know, like, eighty percent of the time. Once the plan is good, and sometimes it takes a little back and forth, I just get Claude to execute. And nowadays, what I find with Opus four point five, I think it started with four point six, it got really good. Once the plan is good, it just stays on track, and it'll just do the thing exactly right, almost every time. And so, you know, before you had to babysit after the plan and before the plan. Now it's just before the plan. So maybe the next thing is you just won't have to babysit. You can just kinda give a prompt, and Claude will figure it out.

    13. GT

      The next step is Claude just speaks to your users directly. [laughing]

    14. JF

      [chuckles] Yeah.

    15. GT

      It just bypasses you entirely.

    16. BC

      It's funny, this is actually the current stuff for us. Our Claudes actually, like, they talk to each other, they talk to our users on Slack, at least, uh, internally, pretty often. Um, my Claude will, like, tweet once in a while.

    17. GT

      No way!

    18. BC

      Um, but I actually, like, delete it [chuckles] because it's just, like, it's a little, like, cheesy.

    19. GT

      Yeah.

    20. BC

      It, it... Like, I don't love the tone.

    21. GT

      What does it want to tweet about?

    22. BC

      Sometimes it'll just, like, respond to someone. 'Cause I always have, like, Cowork running in the background, and it's like, it's the Cowork Claude that really loves to do that 'cause it likes using a browser.

    23. GT

      That's funny.

    24. BC

      Uh, a really common pattern is I ask Claude to build something, it'll look in the code base, uh, it'll see some engineer touch something in the git blame, and then it'll message that engineer on Slack, um, just, like, asking a clarifying question, and then once it gets the answer back, it'll keep going.

    25. JF

      What are some

  13. 28:38 – 30:07

    Tips for founders to build for the future

    1. JF

      tips for founders now on how to build for the future? It sounds like everything is really changing. What are, like, some principles that will stay on and what will change?

    2. BC

      So I think some of these are pretty, are pretty basic, but I think they're even more important now than they were before. Um, so one example is latent demand. Like, I've mentioned it a thousand times. For me, it's just, like, the single biggest idea in product. It's a, it's a thing that no one understands. It's a thing I certainly did not understand my first few startups. And, and the idea is, like, people will only do a thing that they already do. You can't get people to do a new thing. If people are trying to do a thing and you make it easier, that's a good idea. But if, if people are doing a thing and you try to make them do a different thing, they're not gonna do that. And so you just have to make the thing that they're trying to do easier. And I think Claude is gonna get increasingly good at kinda figuring out these kind of product ideas for you, just 'cause it can look at feedback, it can look at debug logs, it can kinda figure this out.

    3. GT

      That's what you mean by plan mode is latent-

    4. SP

      ... demand that people were already like, kind of had their Claude chat window open in a browser and were, like, talking to it to figure out, like, the spec and, and what it should do. And now it's-- that, like, plan mode just became that. You just do it in Claude Code.

    5. BC

      Yeah, yeah. That's it. Some- sometimes what I'll do is I'll just walk around the office on, on our floor, and I'll just kind of stand behind people. [chuckles] Like, I, I'll say, like, "Hi," so it's not weird, and then, um, I'll, I'll just see kind of like how they're using Claude Code. Um, and this is also just something I saw a lot, um, but it also came up in GitHub issues. Like, people were talking about it.

    6. SP

      It seems like so you're surprised how far the terminal has gone and how far it's been pushed.

  14. 30:07 – 30:57

    How much life does the terminal still have?

    1. SP

      Like, how far do you think it has left to go, just given with this world of swarm multiple agents-

    2. BC

      Mm.

    3. SP

      -like, do you think there's gonna be a new, a need for a different UI on top of it?

    4. BC

      It's funny, if you asked me this a year ago, I would've said the terminal has, like, a three-month lifespan, and then we're gonna move on to the next thing. Uh, and you can see us experimenting with this, right? 'Cause Claude Code started in a terminal, but now it's in, you know, it's on web. You can... Like, claude.ai/code. It's in the desktop app. You know, we've had that for, you know, like, three months or six months or something, just in the Code tab. Um, it's in the iOS and Android apps, just, like, in the Code tab. It's in Slack. It's in GitHub. There's VS Code extensions. There's JetBrains extensions. So we're just, like, we're always experimenting with different form factors for this thing to figure out what's the next thing. I've been wrong so far-

    5. SP

      [chuckles]

    6. BC

      -about the lifespan of the CLI, so I'm probably not the person to forecast this. [chuckles]

    7. SP

      [chuckles] What about,

  15. 30:57 – 32:11

    Advice for dev tool founders

    1. SP

      like, your advice to dev tool founders? Like, someone's building a dev tool company today, should they just, like, be building for engineers and humans, or should they be thinking more about, like, what Claude is gonna think and want and build for sort of like the agent?

    2. BC

      The way I would frame it is think about the thing that the model wants to do and figure out how do you make that easier. And that's something that we saw... You know, like, when I first started hacking on Claude Code, I, I realized, like, this thing just wants to use tools. It just wants to interact with the world. And how, how do you, how do you enable that? Well, the way you don't do it is you put it in a box, and you're like: "Here's the API, here's how you interact with me, and here's how you interact with the world." The way you do it is you see what tools it wants to use, you see what it's trying to do, and you enable that, the same way that you do for your users. And so, like, for-- if you're building a dev tools startup, I would think about, like, what is the problem you want to solve for the user? And then when you use-- when you apply the model to solving this problem, what is the thing the model wants to do?

    3. SP

      Hmm.

    4. BC

      And then what is the technical and product solution that serves the weight and demand of both?

    5. SP

      YC's next batch is now taking applications. Got a startup in you? Apply at ycombinator.com/apply. It's never too early, and filling out the app will level up your idea. Okay, back to the video.

    6. JF

      Back in the day,

  16. 32:11 – 35:34

    Claude Code and TypeScript parallels

    1. JF

      more than ten years ago, you were a very heaver- heavy user, and you wrote a book about TypeScript, right? Before TypeScript was cool. This is when everyone was, uh, deep in JavaScript. This is back in early twenty tens, right?

    2. BC

      Yeah, something like that.

    3. JF

      Before TypeScript was a thing, because back then, it's a very weird language. It's not supposed to do a lot of things with being typed in JavaScript, and now it's the right thing, and it feels like Claude Code in the terminal has a lot of parallels with TypeScript at the beginning.

    4. BC

      TypeScript makes a lot of really weird language decisions. So i- if you look at the type system, pretty much anything can be a literal type, for example, and this is, like... This is super weird, 'cause, like, even, like, like, Haskell doesn't even do this. It's just like, it's too extreme, or it has, like, conditional types, which I don't think any language thought of at all.

    5. JF

      It was, like, very strongly typed.

    6. BC

      Yeah, it was very strongly typed. And, and the, the idea was, like, when, you know, like, when Joe Paymer and Anders and the early team was, like, building this thing, the way they built it is we-- okay, we have these teams with these big untyped JavaScript code bases. We have to get types in there, but we're not gonna get engineers to change that, the, the way that they code. You're not gonna get Java- JavaScript people to have, like, you know, fifteen layers of class inheritance like you would a Java programmer.

    7. JF

      Right.

    8. BC

      They're gonna write code the way they're gonna write it. They're, they're gonna use reflection, and they're gonna use mutation, and they're gonna use all these features that traditionally are very, very difficult to type.

    9. JF

      They're a very unsafe type to any, uh, strong functional programmer, really. [chuckles]

    10. BC

      That's right, that's right, that's right. And so the thing that they did, instead of getting people to kinda change the way that they code, they, they built a type system around this, and it was just... It's brilliant because there's all these ideas that no one was thinking about. Even in academia, like, no one thought of a bunch of these ideas. It purely came out of the practice of observing people and seeing how JavaScript programmers want to write code. And so, you know, for, for Claude Code, it, there, there are some ideas that are kinda similar in that, you know, like, you can use it like a Unix utility. You can pipe into it. You can pipe out of it. Um, it- in some ways, it is kinda rigorous in this way, but in al- in almost every other way, it's just the tool that we wanted. Like, I, I built a tool for myself, and then the team built the tool for themselves and then for Anthropic employees and then for users, and it just ends up being really useful. It's not, it's not this, like, principled and academic thing.

    11. JF

      Which I think the, the proof is actually in the results. Now, fast-forward more than fifteen years later, not many code bases are in Haskell, which is more academic, and there's tons of them now in TypeScript, because it's way more practical-

    12. BC

      Right

    13. JF

      ... which is interesting.

    14. BC

      Yeah, it is interesting, right? It's like TypeScript solves a problem.

    15. JF

      I guess one thing that's cool, I don't know how many people know, but the terminal is actually one of the most beautiful terminal apps out there and is actually written with React terminal.

    16. BC

      When I first started building it, you know, like, I, I did front-end engineering for, for a while, so... And I was also, like, a, you know-- I'm, I'm sort of like a hybrid. Like, I, I do, like, design and user research and, you know, write code and all the stuff, and we, we love hiring engineers that are like this. Um, so we just, we love generalists. So for me, it's like: Okay, I'm building a thing for the terminal. I'm actually kind of a shitty Vim user, so, like, how do I build a thing for people like me that, um, you know, are, are gonna be working in a terminal? And I think just the delight is so important, and I feel like at YC, this is something you talk about a lot, right? It's like, build a thing that people love. If the product is useful, but you don't fall in love with it, that's not great. Um, so it kinda has to do both.

  17. 35:34 – 37:36

    Designing for the terminal was hard

    1. BC

      Designing for the terminal honestly has been hard.... right? It's like, uh, it's like eighty by a hundred characters or whatever. You have, like, two hundred and fifty-six colors. You have one font size. You don't have, like, mouse interactions. There's all this stuff you can't do, and there's all these very hard trade-offs. So, like, a little-known thing, for example, is you can actually enable mouse interactions in a terminal, so you can enable, like, clicking and stuff.

    2. GT

      Oh, how do you do that in Claude Code? I've been trying to figure out how to do this. [laughing]

    3. JF

      [laughing]

    4. BC

      We don't, we don't have it in Claude Code 'cause we actually prototyped it a few times, and it felt really bad because the trade-off is you have to virtualize scrolling. And so there's all these weird trade-offs because, like, the way terminals work is, like, there's no DOM, right? It's like there's, like, ANSI escape codes and these kind of weird organically evolved specs since, like, the nineteen sixties or whatever.

    5. JF

      Oh, yeah. It feels like BBSs. It's like a BBS door game.

    6. BC

      Yeah, yeah, yeah.

    7. JF

      Oh, my gosh! [chuckles]

    8. BC

      It's like... That's, like, a great compliment.

    9. JF

      Yeah.

    10. BC

      Yeah, yeah.

    11. JF

      Yeah, it's awesome.

    12. BC

      Like that you feel like you're discovering something.

    13. JF

      Lord of the Red Dragons.

    14. BC

      [chuckles]

    15. JF

      Fantastic. Oh, my God.

    16. BC

      Yeah.

    17. JF

      Yeah.

    18. BC

      But we have- we've had to just, like, discover all these kind of UX principles for building in the terminal 'cause no one really writes about this stuff. And if you look at the big terminal apps of, you know, like the eighties or nineties or two thousands or whatever, they use, like, ed curses, and they have all these, like, windows and things like this, and it just looks kind of, like, janky by modern standards. It just looks too heavy and complicated. And so we had to, like, reinvent a lot. And, you know, for example, something like the terminal spinner, like, just, like, the spinner words, it's gone through probably, I wanna say, like, fifty, maybe a hundred iterations at this point.

    19. JF

      Wow!

    20. BC

      And probably eighty percent of those didn't ship. So we tried it, it didn't feel good, move on to the next one. Try it, didn't feel good, move on to the next one. Uh, and this was, like, sort of one of the amazing things about Claude Code, right? Is like, you can write these prototypes, and you can just do, like, twenty prototypes back-to-back, see which one you like, and then ship that, and the whole thing takes maybe a couple hours. Whereas in the past, what you would have had to do is, like, if you were going to use Origami or Framer or something like this-

    21. JF

      Mm

    22. BC

      ... you build, like, maybe three prototypes. It took, like, two weeks. It just took much, much longer. And so we have this luxury of we have to discover this new thing, we have to build the thing. We don't know what the right endpoint is, but we can iterate there so quickly, and that's what makes it really easy, and that's what lets us build a product that's, like, joyous and that people like to

  18. 37:36 – 40:31

    Other advice for builders

    1. BC

      use.

    2. GT

      Boris, you had other advice for, for builders, and we kept inter- interrupting you 'cause we have so many questions. [laughing]

    3. JF

      [laughing]

    4. BC

      I would say, um... So, okay, so maybe two pieces of advice that are kinda weird because it's, like, about building for the model. So one is, uh, don't build for the model of today, build for the model of six months from now. This is, like, sort of weird, right? Because, like, you can't find PMF if the product doesn't work, but actually, this is the thing that you should do, because otherwise, what will happen is you spend a bunch of work, you find PMF for the product right now, and then you're just gonna get leapfrogged by someone else, um, because they're building for the next model, and a new model comes out every few months. Use the model, feel out the boundary of what it can do, and then build for the model that you think will be the model maybe six months from now. I think the second thing is, um, you know, actually, in the, in the Claude Code, we're in the Claude Code area, where we sit- we have a framed copy of The Bitter Lesson on the wall. Um, and this is this, like, Rich Sutton, uh, paper that, like, everyone should read it if, if you haven't. Uh, and the, the idea is the more general model will always beat the more specific model. And there's a lot of corollaries to this, but essentially, what it boils down to is never bet against the model. Uh, and so this is just, like, a thing to... That, that we always think about, where we could build a feature into Claude Code, we could make it better as a product, and we call this scaffolding. It's all this code that's not the model itself. But we could also just wait, like, a couple months, and the model can probably just do the thing instead. Um, and there's always this trade-off, right? It's like engineering work now, and you can kinda extend the capability a little bit, maybe ten, twenty percent or whatever, in whatever domain on this, like, you know, like, the spider chart of what you're trying to extend. Um, or you can just wait, and the next model will do it. So just alway- always think in terms of this trade-off. Where, where do you actually wanna invest? And assume that whatever the scaffolding is, it's just tech debt.

    5. JF

      How often do you rewrite the code base of, uh, Claude Code? Is this every six months with this, with this [chuckles] first model?

    6. GT

      Yeah, is there scaffolding that you've deleted 'cause you don't need it anymore 'cause the model just improved?

    7. BC

      Oh, so much. Yeah, like, uh, all of Claude Code has just been written and rewritten and rewritten and rewritten over and over and over. We unship tools every couple weeks. We add new tools every couple weeks. There's no part of Claude Code that was around six months ago. It's just constantly rewritten.

    8. JF

      Would you say that most of the code base for our current Claude Code is only, say, eighty percent of it, is only less than a couple months old?

    9. BC

      Yeah, definitely. It might, it might even be, like, less than... Yeah, maybe, like, a couple months. That, that feels about right.

    10. JF

      So it's like the life cycle of code now, that's another alpha, is expecting it to be the shelf life to be just a couple months-

    11. BC

      Yeah

    12. JF

      ... for the best founders.

    13. BC

      Yeah.

    14. GT

      Do you see, uh, Steve Yegge's, uh, post about, uh, how awesome working at Anthropic is? And I think there's a line in there that says that, "An Anthropic engineer, uh, currently averages one thousand X more productivity than a Google engineer at Google's peak," which is really an insane number, [chuckles] honestly, like one thousand X. Like, you know, we're- three years ago, we were still talking about ten X engineers. Now we're talking about [laughing] a thousand X-

    15. BC

      [laughing]

    16. JF

      [laughing]

    17. GT

      ... on top of a Google engineer in their prime? Like, this is unbelievable, honestly.

    18. BC

      Yeah,

  19. 40:31 – 41:36

    Productivity per engineer

    1. BC

      I mean, internally, if you, if you look at, like, technical employees, they all use Claude Code every day. Um, and even non-technical employees, I think, like, half the sales team uses Claude Code.

    2. GT

      Mm.

    3. BC

      Um, they, they've started switching to Cowork 'cause it's a little easier to use. It has, like, a VM, so it's a little bit safer. But yeah, we actually, we just pulled the stat, and the, the- I think the team doubled in size last year, but productivity per engineer grew something like seventy percent.

    4. GT

      As measured by?

    5. BC

      Just, like, the st- simplest, stupidest measure, pull requests.

    6. GT

      Okay.

    7. BC

      Um, but we also kinda cross-check that against, like, commits and, like, uh, the lifetime of commits and things like this.

    8. GT

      Mm.

    9. BC

      And since Claude Code came out, productivity per engineer at Anthropic has grown a hundred and fifty percent.

    10. GT

      Oh, my God.

    11. BC

      Um, and this is crazy because I want- in my old life, I was responsible for code quality at Meta.

    12. JF

      Hmm.

    13. BC

      Um, and I was responsible for the quality of all of our code bases across every product, across, like, you know, Facebook, Instagram, WhatsApp, whatever. And one of the things that the team worked on was improving productivity. And back then, seeing a gain of something like two percent in productivity, that was, like, a year of work by hundreds of people. And so this, like, a hundred percent, this is just, like, unheard of, just completely unheard of.

    14. GT

      What drove you

  20. 41:36 – 44:46

    Why Boris chose to join Anthropic

    1. GT

      to come over to Anthropic? I mean, basically, as a builder, you could go anywhere. What was the moment that made you say, like: "Actually, this is the set of people," or, "This is the approach?"

    2. BC

      I was living in rural Japan, and I was opening up-... Hacker News every morning, and I was reading the news, and, uh, it was all- it just started to be, like, AI stuff at some point. And, uh, I started to use some of these early products, and, uh, I remember, like, the first couple times that I used it, I was just like, m- it, it just took my breath away. That was, like, very cheesy to say, but that was actually, that was actually the feeling. Like, it, it was just, like, it was amazing. Like, as a, as, as a builder, I've just never kind of felt, felt this feeling, like, using these very, very early products. That was, like, in the Claude two days or, you know, something like that. And so I, I just talked to- started talking to friends at labs, um, just to kinda see what was going on. Um, and, uh, I met Ben Mann, who's one of the founders at, uh, at Anthropic, and, uh, he just immediately won me over. Um, and as soon as I met kind of the rest of the team at And, they just won me over, and I think, I think probably in two ways. So one is it operates as a research lab. Um, so the product work was teeny, teeny, tiny. It's really all about building a safe model. That's all that matters. Um, and so this idea of just being very close to the model and being very close to development and being not the most important thing, because the product isn't anymore, it's just the model is the thing that's the most important. Um, that really resonated with me after building product for many years. And then the second thing was just how mission-driven it is. Um, like I'm, I'm a huge sci-fi reader. My bookshelf is just, like, filled with sci-fi, and so, like, I just know how bad this can go.

    3. GT

      Mm-hmm.

    4. BC

      And when I kinda think about what's gonna happen this year, it, you know, it's gonna be totally insane. And in the worst case, it can go very, very bad. Um, and so I just wanted to be at a place that really understood that and kinda really internalized that. And at And, you know, like, if you overhear conversations in the lunchroom or in the hallway, people are talking about AI safety. This is really the thing that everyone cares about more than anything. Um, and so I just wanted to be in a place like that. I, I know, I know for me personally, the mission is just so important.

    5. GT

      What is gonna happen this year? [laughing]

    6. BC

      [chuckles] Okay, so i- if you think back, like, six months ago and, uh, kinda what are the predictions that people are making? So Dario predicted that ninety percent of the code at, at Anthropic would be, would be written by Claude. This is true. Um, for me personally, it's been one hundred percent for, like, since Opus Four point five. Um, I just-- I uninstalled my IDE. I don't edit a single line of code by hand. It's just a hundred percent Claude Code and Opus. Um, and, you know, I land, you know, like, twenty PRs a day, every day. If you look at Anthropic overall, it ranges between, like, seventy to ninety percent, uh, you know, depending on the team. For a lot of teams, it's also, like, one hundred percent. For a lot of people, it's one hundred percent. And I remember making this prediction back in May, when we GA'd Claude Code, that you wouldn't need an IDE to code anymore. Uh, and it was totally crazy to say. I feel like people in the audience gasped- [chuckles]

    7. GT

      [chuckles]

    8. BC

      -because it was such, like, a silly prediction at the time. But really, all it is is, like, you just, like, trace the, you know, the exponential. And this is just, like, so deep in, you know, the DNA at And, 'cause, like, you know, three of our founders were co-authors of the Scaling Laws paper. They kinda, they saw this very early. And so this is just like tracing the exponential, this is what's gonna happen, and yes, that happened. So

  21. 44:46 – 46:22

    How coding will change

    1. BC

      continuing to trace the exponential, I think what will happen is coding will be generally solved for everyone. Um, and I think today, coding is practically solved, you know, for me, and I think it'll be the case for everyone, um, you know, re- regardless of domain. I think we're gonna start to see the title software engineer go away, and I think it's just gonna be maybe builder, maybe product manager. Maybe we'll keep the title as kind of a vestigial thing, but the work that people do, it's not just gonna be coding, it's software engineers are also gonna be writing specs. They're gonna be talking to users. Like, this thing that we're starting to see right now in our team, where engineers are very much generalists, and every single function on our team codes, like our PMs code, our designers code, our EM codes, our, um, like, everyone, our, our finance guy codes. Like, everyone on our team codes. We're gonna start to see this everywhere. So this is sort of, uh, this is kind of like the lower bound if we just continue with the trend. The upper bound, I think, is a lot scarier, um, and this is something like, you know, we hit ASL-4. Um, and the, you know, at Anthropic, we talked about the safety levels. ASL-3 is where the models are right now. ASL-4 is the model is recursively self-improving. Um, and so if this happens, essentially, we have to meet a bunch of criteria before we can release a model. And so the, the extreme is that, you know, this happens, um, or there is some kind of catastrophic misuse. Like, people are using the model to design bio viruses, design zero-days, stuff like this. Um, and this is something that we're really, really actively working on, so that doesn't happen. I think, uh, it, it's just been-- honestly, it's just been, like, so exciting and humbling, like, seeing how people are using Claude Code. Like, uh, you know, I just wanted to build a cool thing, and it ended up being really useful, uh, and that was so surprising and so exciting.

    2. GT

      My impression from

  22. 46:22 – 50:10

    Outro

    1. GT

      Twitter or just the outside is basically everyone went away over the holidays and then, like, found out about Claude Code, and it's just been crazy ever since. But is that how it was for you at, like, internally? Did you-- were you having, like, a nice Christmas break, and [chuckles] then came back, and you're like, "What happened?"

    2. BC

      Well, actually, for all of December, I was traveling around, uh, and I, I took a coding vacation, so we were kinda traveling around, and I was just, like, coding every day, so that was really nice. Uh, and then I also started to use Twitter at the time 'cause, like, I, I worked on Threads back then-

    3. GT

      Yeah

    4. BC

      ... way back when, so I've been a Threads user for a while. So I just, like, tried to see kind of like other platforms where people are. Yeah, I think for a lot of people, they kinda discover-- That was the moment where they discovered Opus Four point five. I kind of already knew.

    5. GT

      Mm-hmm.

    6. BC

      Uh, and internally, Claude Code's just been on this, like, exponential tear for many, many months now, so that just, like, it, it became even more steep. That's what we saw. And if you look at Claude Code now, you know, there was some stat from Mercury that, like, seventy percent of startups are, you know, choosing Claude as their model of choice. There was some other stat from, like, Semi Analysis, that four percent of all public commits are made by Claude Code, um, like, of all code written everywhere.

    7. GT

      I saw that.

    8. BC

      All the companies, you know, use Claude Code, from, like, the biggest companies to kinda, you know, smaller startups. You know, like, it, it wrote, it, it plotted the course for Perseverance, like, for, like, the Mars rover. This is just, like, this is the coolest thing for me, and we, like, we even printed posters-

    9. GT

      Mm-hmm

    10. BC

      ... 'cause the team was like: "Wow, this is just, like, so cool that NASA chooses to use this thing." So yeah, it's just, like, it's humbling, um, but it also feels like the very beginning.

    11. GT

      What's the sort of interaction between, uh, Claude Code and then Cowork? Like, you know, was it-

    12. GT

      ... a fork of Claude Code? Was it like you had Claude Code look at the Claude Code code and say, "Let's make a new spec for non-technical people that, you know, keeps all the lessons," and then, you know, it sort of went off for a couple days and did that? What's the genesis of that and, you know, where do you think that goes?

    13. BC

      This is gonna be, like, my fifth time using the word wait and demand. [chuckles]

    14. GT

      Yeah. [chuckles]

    15. BC

      It was just that. I mean, like, we, we were looking at Twitter, and there was, like, that one guy that was using Claude Code to, like, monitor his tomato plants.

    16. GT

      Mm-hmm.

    17. BC

      Uh, there was, like, this other person that was using it to, like, recover wedding photos off of a corrupted hard drive. There were people that using it for, like, uh, for finance. When we looked internally at Anthropic, every designer is using it. All- the entire finance team at this point is using it. The entire data science team is using it, not for coding. People are jumping over hoops to install a thing in the terminal so that they can use this. So we knew for a while that we wanted to build something, and so we were experimenting with a bunch of different ideas, and the thing that kinda took off was just, you know, a little Claude Code wrapper in a GUI in the desktop app. And that, that's all it is, it's just Claude Code under the hood. It's the same agent.

    18. GT

      Oh, wow.

    19. BC

      Um, and, uh, Felix and the team... And Felix was an early Electron contributor. He kinda knows that stack really well, and he was hacking on various ideas. And, uh, they, they built it in, I think, something like 10 days. It was, it was just, like, 100% written by Claude Code, uh, and it just felt ready to release. There was a lot of stuff that we had to build for non-technical users, so it's a little bit different than a technical audience. Uh, it runs in a... All the code runs in a virtual machine. Uh, there's a lot of dele- uh, protections for deletion and things like this. There's a lot of permission prompting and kinda other guardrails for users. Um, but yeah, it was honestly pretty obvious.

    20. GT

      Boris, thank you so much for making something that, uh, is taking away all my sleep, but in return, it's making me feel creator mode again, sort of founder mode again. It's been an exhilarating three weeks. I, like, can't believe I waited that long since November to actually get into it. Thank you so much for being with us. Thank you for building what you're building.

    21. BC

      Yeah. Thanks for having me, and, uh, sandbox? [laughing]

    22. GT

      [laughing] Sounds good.

    23. SP

      [upbeat music]

Episode duration: 50:10

Install uListen for AI-powered chat & search across the full episode — Get Full Transcript

Transcript of episode PQU9o_5rHC4

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome