Claude Code head Boris Cherny: Why he ships 30 PRs a day

1. 0:00 – 3:45

   Introduction to Boris and Claude Code

   1. BCBoris Cherny0:00

      A hundred percent of my code is written by Claude Code. I have not edited a single line by hand since November. Every day, I ship ten, twenty, thirty pull requests. So, like, at the moment, I have, like, five agents running.

   2. LRLenny Rachitsky0:10

      While we're recording this?

   3. BCBoris Cherny0:11

      Yeah, yeah, yeah.

   4. LRLenny Rachitsky0:12

      Do you miss writing code?

   5. BCBoris Cherny0:13

      I have never enjoyed coding as much as I do today, because I don't have to deal with all the minutiae. Productivity per engineer has increased two hundred percent. There's always this question: Should I learn to code? In a year or two, it's not gonna matter. Coding is largely solved. I imagine a world where everyone is able to program. Anyone can just build software anytime.

   6. LRLenny Rachitsky0:29

      What's the next big shift to how software is written?

   7. BCBoris Cherny0:32

      Claude is starting to come up with ideas. It's looking through feedback, it's looking at bug reports, it's looking at telemetry for bug fixes and things to ship. A little more like a coworker or something like that.

   8. LRLenny Rachitsky0:41

      A lot of people listening to this are product managers, and [chuckles] they're probably sweating.

   9. BCBoris Cherny0:44

      I think by the end of the year, everyone's gonna be a product manager, and everyone codes. The title software engineer is gonna start to go away. It's just gonna be replaced by builder, and it's gonna be painful for a lot of people.

   10. LRLenny Rachitsky0:53

       [gentle music] Today, my guest is Boris Cherny, head of Claude Code at Anthropic. It is hard to describe the impact that Claude Code has had on the world. Around the time this episode comes out will be the one-year anniversary of Claude Code, and in that short time, it has completely transformed the job of a software engineer, and it is now starting to transform the jobs of many other functions in tech, which we talk about. Claude Code itself is also a massive driver of Anthropic's overall growth over the past year. They just raised a round at over three hundred and fifty billion dollars, and as Boris mentions, the growth of Claude Code itself is still accelerating. Just in the past month, their daily active users has doubled. Boris is also just a really interesting, thoughtful, deep-thinking human, and during this conversation, we discover we were born in the same city in Ukraine. That is so funny. I had no idea. A huge thank you to Ben Mann, Jenny Wen, and Mike Krieger for suggesting topics for this conversation. Don't forget to check out lennysproductpass.com for an incredible set of deals available exclusively to Lenny's Newsletter subscribers. Let's get into it after a short word from our wonderful sponsors. Today's episode is brought to you by DX, the developer intelligence platform designed by leading researchers. To thrive in the AI era, organizations need to adapt quickly, but many organization leaders struggle to answer pressing questions like: Which tools are working? How are they being used? What's actually driving value? DX provides the data and insights that leaders need to navigate this shift. With DX, companies like Dropbox, Booking.com, Adyen, and Intercom get a deep understanding of how AI is providing value to their developers and what impact AI is having on engineering productivity. To learn more, visit DX's website at getdx.com/lenny. That's getdx.com/lenny. Applications break in all kinds of ways: crashes, slowdowns, regressions, and the stuff that you only see once real users show up. Sentry catches it all. See what happened, where, and why, down to the commit that introduced the error, the developer who shipped it, and the exact line of code all in one connected view. I've definitely tried the five tabs and Slack thread approach to debugging. This is better. Sentry shows you how the request moved, what ran, what slowed down, and what users saw. Seer, Sentry's AI debugging agent, takes it from there. It uses all of that Sentry context to tell you the root cause, suggest a fix, and even opens a PR for you. It also reviews your PRs and flags any breaking changes with fixes ready to go. Try Sentry and Seer for free at sentry.io/lenny, and use code Lenny for one hundred dollars in Sentry credits. That's S-E-N-T-R-Y.io/lenny.

2. 3:45 – 5:35

   Why Boris briefly left Anthropic for Cursor (and what brought him back)

   1. LRLenny Rachitsky3:46

      [gentle music] Boris, thank you so much for being here, and welcome to the podcast.

   2. BCBoris Cherny3:54

      Yeah, thanks for having me on.

   3. LRLenny Rachitsky3:55

      I wanna start with a, a spicy question. About six months ago, I don't know if people even remember this, you actually left Anthropic, you joined Cursor, and then two weeks later, you went back to Anthropic. What happened there? I don't think I've ever heard the actual story.

   4. BCBoris Cherny4:10

      [chuckles] It's the fastest job change that I've ever had. [chuckles] Um, I joined Cursor because I'm a big fan of the product, and honestly, I met the team, and I was just really impressed. Uh, they're an awesome team. Uh, I still, I still think they're awesome, and they're just building really cool stuff and kinda... They, they saw where AI coding was going, I think, before a lot of people did. So the, the idea of building good product was just very exciting for me. I think as soon as I got there, what I started to realize is what I really missed about Ant was the mission, and that's actually what originally drove me to Ant also. 'Cause, uh, but before I joined Anthropic, I was, you know, I was working in big tech, and then I was... At, at some point, I wanted to work at a, at a lab to just help shape the future of this crazy thing that, that we're building in some way. And the thing that drew me to Anthropic was the mission, and it was, you know, it's all about safety. And when you talk to people at Anthropic, just, like, find someone in the hallway, if you ask them why they're here, the answer is always gonna be safety. Um, and so the- this kind of, like, mission-drivenness just really, really resonated with me, and I just know personally, it's something I need in order to be happy. Um, and I... That's just a thing that I really missed, and I found that, you know, whatever the work might be, no matter how exciting, even if it's building a really cool product, it's just not really a substitute for that. Um, so for me, it was actually a, um, it was pretty obvious that, that I was missing that pretty quick.

3. 5:35 – 8:41

   One year of Claude Code

   1. LRLenny Rachitsky5:35

      Okay, so let me follow the thread of just coming back to Anthropic and the work you've done there. This podcast is gonna come out around the year anniversary of launching Claude Code, so I'm gonna spend a little time just reflecting on the impact that you've had. There's, um, this report that recently came out that I'm sure you saw by Semianalysis, that showed that four percent of all GitHub commits are authored by Claude Code now, and they predicted it'll be a, a fifth of all code commits on GitHub by the end of the year. The way they put it is, "While we blinked, AI consumed all software development."... the day that we're recording this, Spotify just put out this, uh, headline that their best developers haven't written a line of code since December, thanks to AI. More and more of the most advanced senior engineers, including you, are sharing the fact that you don't write code anymore, that it's all AI-generated, and many aren't even looking at code anymore, is how far we've gotten. In large part, thanks to this little project that you started and that your team has scaled over the past year. I'm curious just to hear your reflections on, on this past year and the impact that your work has had.

   2. BCBoris Cherny6:42

      These numbers are just totally crazy, right? Like four, four percent of all commits in the world is just way more than I imagined, and, uh, like, like you said, it still feels like the starting point. Um, these are also just public commits, so we actually think if you look at private repositories, it's quite a bit higher than that. And I, I think the craziest thing for me isn't even the number that we're at right now, but the pace at which we're growing. Because if you look at Claude Code's growth rate kind of across any metric, it's continuing to accelerate. Um, so it's not just going up, it's going up faster and faster. When I first started Claude Code, it, it was just gonna be a-- like, it, it was just supposed to be a little hack. Um, you know, we, we broadly knew at Anthropic that we wanted to get a-- we wanted to ship some kind of coding product. And, you know, for Anthropic, for a long time, we were building the models in this way that kind of fit our mental model of the way that we build safe AGI, where the model starts by being really good at coding, then it gets really good at tool use, then it gets really good at computer use. Roughly, this is, like, the trajectory. Uh, and you know, we've been working on this for a long time, and when you look at the team that I started on, it was called the Anthropic Labs team, uh, and actually, Mike Krieger and, you know, Ben Mann, they just kicked this team off again, uh, for kind of round two. The team built some pretty cool stuff. So we built Claude Code, we built MCP, we built the desktop app. So you can kind of see the seeds of this idea. You know, like, it's coding, then it's tool use, then it's computer use. And the reason this matters for Anthropic is, uh, because of safety. It's kind of, again, just back to that AI is getting more and more powerful. It's getting more and more capable. The thing that's happened in the last year is that for, at least for engineers, the AI doesn't just write the code. It, it's not just a conversation partner, but it actually uses tools. It acts in the world. Um, and I think now with Cowork, we're starting to see the transition for non-technical folks also. Um, for a lot of people that use conversational AI, this might be the first time that they're using a thing that actually acts. It can actually use your Gmail, it can use your Slack, it can do all these things for you, and it's quite good at it. Um, and it's only gonna get better

4. 8:41 – 13:29

   The origin story of Claude Code

   1. BCBoris Cherny8:41

      from here. So I think for Anthropic, for a long time, there was this feeling that we wanted to build something, but it wasn't obvious what. And so, uh, when I joined Anth, I spent one month kind of hacking and, you know, built a bunch of, like, weird prototypes. Most of them didn't ship and, you know, weren't even close to shipping. It was just kind of understanding the boundaries of what the model can do. Then I spent a month doing post-training, um, so to understand kind of the research side of it. And I, I think, honestly, that's just for me as an engineer, I find that to do good work, you really have to understand the layer under the layer at which you work. And with traditional engineering work, you know, if you're working on product, you want to understand the infrastructure, the runtime, the virtual machine, the language, kind of whatever that is, the system that you're building on. But, uh, yeah, if you work, if you work in AI, you just really have to understand the model to some degree, to, to do good work. So I took a little detour to do that, and then I came back and just started prototyping what eventually became Claude Code. Uh, and the very first version of it, I have, like, a-- there's, like, a video recording of this somewhere, 'cause I recorded this demo and I posted it. It was called Claude CLI back then, and I just k- kind of showed off how it used a few tools, and the shocking thing for me was that I gave it a bash tool, and, uh, it just was able to use that to write code to tell me what music I'm listening to when I asked it, like: "What music am I listening to?" And this is the craziest thing, right? 'Cause it's like there's no way-- I, I didn't instruct the model to say, you know, "Use, you know, this tool for this," or kind of do whatever. The model was given this tool, and it figured out how to use it to answer this question that I had, that I wasn't even sure if it could answer: What music am I listening to? And so I, I, I started prototyping this a little bit more. Um, I made a post about it, and I announced it internally, and it got two likes. That's the- [chuckles] that was like, that was, like, extent of the reaction at the time. 'Cause I think people internally, you know, like, when you think of coding tools, you think of like, you think of IDEs, you think about kind of all these pretty sophisticated environments. No one thought that this thing could be terminal-based. Um, that's sort of a weird way to design it, and that wasn't really the intention. But, uh, you know, from the start, I built it in a terminal because, you know, for the first couple months, it was just me, so it was just the easiest way to build. Uh, and for me, this is actually a pretty important product lesson, right? It's like you wanna under-resource things a little bit at the start. Then we started thinking about what other form factors we should build, and we actually decided to stick with the terminal for a while. And the biggest reason was the model is improving so quickly. We felt that there wasn't really another form factor that could keep up with it. And honestly, this was just me kind of, like, struggling with kind of like, what should we build? You know, like, for the last year, Claude Code has just been all I think about, and so just, like, late at night, this is just something I was thinking about, like, okay, the model is continuing to improve. What do we do? How can we possibly keep up? And the terminal was honestly just the only idea that I had. And, uh, yeah, it ended up catching on. After, after I released it, pretty quickly, it became a hit at Anthropic, and, you know, the, the daily active users just went vertical. And it really early on, actually, before I launched it, Ben Mann, uh, nudged me to make a DAU chart, and I was like: "You know, it's, like, kind of early. Maybe, you know, should we really do it right now?" And he was like: "Yeah." And so the, the chart just went vertical pretty immediately. Uh, and then in February, we released it externally. Actually, something that people don't really remember is Claude Code was not initially a hit when we released it. It, it got a bunch of users. There was a lot of early adopters that got it immediately, but it actually took many months for everyone to really understand what this thing is. Just again, it's like it's just so different.... And when I think about it, kind of part of the reason Claude Code works is this idea of latent demand, where we bring the tool to where people are, and it makes existing workflows a little bit easier. But also because it's, it's in a terminal, it's, like, a little surprising. It's a little alien in this way. So you have to, you have to kinda be open-minded, and you have to learn to use it. And of, of course, now, you know, Claude Code is available, you know, in the iOS and Android Claude app. It's available in the desktop app. It's available on the website. It's available as IDE extensions in Slack and GitHub, you know, all of these places where engineers are, it's a little more familiar. But that wasn't the starting point. So yeah, I mean, at the beginning, it was kind of a surprise that this thing was even useful. And, uh, you know, as the team grew, as the product grew, as it started to become more and more useful to people, just people around the world from, you know, small startups to the biggest FAANG companies started using it, and they started giving feedback. And I think just reflecting back, it, it's been such a humbling experience, 'cause we just, we keep learning from our users. And just the most exciting thing is, like, you know, none of us really know what we're doing, um, and we're just trying to figure it out along with everyone else, and the single best signal for that is just feedback from users. Um, so that's just been the best. I've be- I've been surprised so many times.

5. 13:29 – 15:01

   How fast AI is transforming software development

   1. LRLenny Rachitsky13:29

      It's incredible how fast something can change in today's world. You launched this a year ago, and it wasn't the first time people could use AI to code, but, uh, in a year, the entire profession of software engineering has dramatically changed. Like, there was all these predictions, "Oh, AI is gonna be written-- hundred percent AI is, uh, or code is gonna be written by AI." Everyone's like: No, that's crazy. What are you talking about?

   2. BCBoris Cherny13:51

      Mm.

   3. LRLenny Rachitsky13:51

      And now it's like, oh, of course, it's happening exactly as they said. It's just so things move so fast and change so fast now.

   4. BCBoris Cherny13:58

      Yeah, it's really fast. Back at, uh, back at Code with Claude back in May, that was, like, our first, uh, you know, like, developer conference that we did as Anthropic. Um, I did a short talk, and the, in the Q&A after the talk, people were asking: What are your predictions for the end of the year? And my prediction back in May of twenty twenty-five was, by the end of the year, you might not need a IDE to code anymore, and we're gonna start to see engineers not doing this. And I remember the room, like, audibly gasped. [chuckles] It was such a crazy prediction. But I think, like, at, at Anthropic, like, this is just the way-- the way we think about things is exponentials, and this is, like, very deep in the DNA. Like, if you look at our co-founders, like, three of them were the first three authors on the scaling laws paper. Um, so we really just think in exponentials, and if you kinda look at the exponential of the percent of code that was written by Claude at that point, if you just trace the line, it's pretty obvious we're gonna cross a hundred percent by the end of the year, even if it just does not match intuition at all. Um, and so all I did was trace the line, and, yeah, in November, that, you know, that happened for me personally, and that's been the case since, and we're starting to see that for a lot of different customers, too.

6. 15:01 – 16:17

   The importance of experimentation in AI innovation

   1. LRLenny Rachitsky15:01

      I thought was really interesting what you just shared there about kind of the journey is this kind of idea of just playing around and seeing what happens. This came up, comes up with OpenClaus a lot, just like Peter was playing around, and just, like, a thing happened. And it feels like that's a central kind of ingredient to a lot of the biggest innovations in AI, is people just sitting around trying stuff to... Pushing the models further than most other people.

   2. BCBoris Cherny15:22

      I mean, this is the thing about innovation, right? Like, you can't, uh, you can't force it. There's no roadmap for innovation. Um, you just have to give people space. You have to give them, maybe the word is, like, safety. So it's like psychological safety, that it's okay to fail. It's okay if eighty percent of the ideas are bad. Um, you also have to hold them accountable a bit, so if the idea is bad, you, you know, you cut your losses and move on to the next idea instead of investing more. Uh, in the early days of Claude Code, I had no idea that this thing would be useful at all, 'cause [chuckles] even in February when we released it, it was writing maybe, I don't know, like, twenty percent of my code, not more. And even in May, it was writing maybe thirty percent. I was still using, you know, Cursor for most of my code, and it only crossed a hundred percent in November, so it took a while. But even from the earliest day, it just felt like I was onto something, and I was just spending, like, every night, every weekend hacking on this, and luckily, my, you know, my wife was very supportive. Um, but it, it, it just felt like it was onto something. It wasn't obvious what, and, and sometimes, you know, you find a thread, you just have to pull on

7. 16:17 – 17:32

   Boris’s current coding workflow (100% AI-written)

   1. BCBoris Cherny16:17

      it.

   2. LRLenny Rachitsky16:17

      So at this point, one hundred percent of your code is written by Claude Code. Is that, is that kind of the current state of your coding?

   3. BCBoris Cherny16:23

      Yeah, so a hundred percent of my code is written by Claude Code. Um, I'm a fairly prolific coder, um, and th- this has been the case even when I worked back at Instagram. I was, like, one of the top few most productive engineers. Um, and that's actually, that's still the case, uh, here at Anthropic.

   4. LRLenny Rachitsky16:38

      Wow!

   5. BCBoris Cherny16:38

      Um, so-

   6. LRLenny Rachitsky16:38

      Even as head of, [chuckles] head of the team.

   7. BCBoris Cherny16:41

      Yeah, yeah. Do still do a lot of coding. Um, and so every, you know, every day I ship, like, ten, twenty, thirty pull requests, something like that.

   8. LRLenny Rachitsky16:47

      Every day?

   9. BCBoris Cherny16:48

      A hun-- every day.

   10. LRLenny Rachitsky16:49

       [chuckles]

   11. BCBoris Cherny16:49

       Yeah.

   12. LRLenny Rachitsky16:50

       Good God.

   13. BCBoris Cherny16:50

       Uh, hundred percent written by Claude Code. I have not edited a single line by hand since, uh, November, and yeah, that, that's been it. I do look at the code, so I, I don't think we're kind of at the point yet where you can be totally hands-off, especially when there's a lot of people, you know, like, running the program. You have to make sure that it's correct, you have to make sure it's safe, and so on. Um, and then we also have Claude doing automatic code review for everything. Um, so here at Anthropic, Claude reviews a hundred percent of pull requests. Um, there's still a layer of, like, human review after it, but you kind of, like, you still do want some of these checkpoints. Like, you still want a human looking at the code, um, unless it's, like, pure prototype code that, you know, it's not gonna run, it's not gonna run anywhere. It's just a prototype.

8. 17:32 – 22:24

   The next frontier

   1. LRLenny Rachitsky17:32

      What's kind of the next frontier? So at this point, a hundred percent of your code is being written by AI. This is clearly where everyone is going in software engineering. That felt like a crazy milestone. Now, it's just like, of course, this is the world now. What's, what's kind of the next big shift to how software is written that either your team's already operating in or you think will he- head towards?

   2. BCBoris Cherny17:54

      I think something that's happening right now is Claude is starting to come up with ideas. Um, so Claude is looking through feedback, it's, uh, looking at bug reports, it's looking at, um, you know, like, telemetry and, and things like this, and it's starting to come up with ideas for bug fixes and things to ship. So it's just starting to get a little more, um, you know, like a little more like a coworker or something like that.... I think the second thing is we're starting to branch out of coding a little bit. So I think at this point it's safe to say that coding is largely solved, at least for the kinds of programming that I do, it's just a solved problem, because Claude can do it. And so now we're starting to think about, okay, like, what's next? What's beyond this? There's a lot of things that are kinda adjacent to coding, um, and I think this is gonna be coming, but also just, you know, general tasks. You know, like, and I use Cowork every day now to do all sorts of things that are just not related to coding at all, and just to do it automatically. Like for example, I had to pay a parking ticket the other day. I just had Cowork do it. Um, all of my project management for the team, uh, Cowork does all of it. It's like syncing stuff between spreadsheets and messaging people on Slack and email and all this kind of stuff. So I, I think the frontier is something like this, and I, I don't think it's coding, 'cause, because I think coding is, you know, it's pretty much solved. And over the next few months, I think what we're gonna see is just across the industry, it's gonna become increasingly solved, you know, for every kind of code base, every tech stack that people work on.

   3. LRLenny Rachitsky19:14

      This idea of helping you come up with what to work on is so interesting. A lot of people listening to this are product managers, and [chuckles] they're probably sweating. How do you use Claude for this? Do you just talk to it? Is there anything clever you've come up with to help you use it to come up with what to build?

   4. BCBoris Cherny19:30

      Honestly, the simplest thing is, like, open Claude Code or Cowork and point it at a Slack thread. Um, you know, like, for us, we have this channel that- that's all the internal feedback about Claude Code. Since we first released it, even in like 2024 internally, it's just been this fire hose of feedback. Um, and it's the best. And, like, in the early days, what I would do is any time that someone sends feedback, I would just go in, and I would fix every single thing as fast as I possibly could. So, like, within a minute, within five minutes or whatever. And this just really fast feedback cycle, it encourages people to give more and more feedback. It's just so important, 'cause it makes them feel heard. 'Cause, you know, like, usually when you use a product, you give feedback, it just goes into a black hole somewhere, and then you don't get feedback again. So if you make people feel heard, then they wanna contribute, and they, they wanna help make the thing better. Um, and so now I kinda do the same thing, but Claude honestly does a lot of the work. So I point it at the channel, and it's like: Okay, here's a few things that I can do. I just put up a couple of PRs. Want to take a look at them? And I'm like, "Yeah."

   5. LRLenny Rachitsky20:25

      Have you noticed that it is getting much better at this? Because this is kinda the holy grail. Right now, it's like, cool, building, solved. Code review became kind of the next bottleneck. With all these PRs, who's gonna review them all? The next big o- open question is just like, okay, now we need a... Now, now humans are necessary for figuring out what to build, what to prioritize, and you're saying that's where Claude Code is starting to help you. Has it, has it gotten a lot better with, like, say, O- Opus four six, or what's been the trajectory there?

   6. BCBoris Cherny20:50

      Yeah, yeah, it's improved a lot. Um, I, I think some of it is kinda like training that we do specific to coding. Um, so, you know, obviously, you know, b- best coding model in the world, and, you know, it's getting better and better. Like, four point six is just incredible. But also, actually, a lot of the training that we do outside of coding translates pretty well, too. So there is this kinda like transfer where you teach the model to do, you know, X, and then it kinda gets better at Y. Um, yeah, and it... The, the gains have just been insane. Like, at Anthropic, over the last year, like, since we introduced Claude Code, we've probably, I don't know the exact number, we're probably like four X the engineering team or something like this. But productivity per engineer has increased two hundred percent in terms of, like, pull requests. And, like, this number is just crazy for anyone that actually works in the space and works on dev productivity. 'Cause back in a previous life, I was at Meta, and, you know, one of my responsibilities was code quality for the company. So this is like the, all of our code bases, that, that was my responsibility, like Facebook, Instagram, WhatsApp, all this stuff. Um, and a lot of that was about productivity, because if you make the code higher quality, then engineers are more productive. And things that we saw is, you know, in a year, with hundreds of engineers working on it, you would see a gain of, like, a few percentage points of productivity, something like this. [chuckles] Um, and so nowadays, seeing these gains of just hundreds of percentage points, it's, it's just absolutely insane.

   7. LRLenny Rachitsky22:06

      What's also insane is just how normalized this has all been. Like, we hear these numbers, like, of course, AI is doing this to us. It's just, it's so unprecedented, the amount of change that is happening to software development, to building products, to just this, the world of tech. It's just, like, so easy to get used to it, but it's important to recognize this

9. 22:24 – 24:02

   The downside of rapid innovation

   1. LRLenny Rachitsky22:24

      is crazy.

   2. BCBoris Cherny22:25

      This is something, like, I have to remind myself once in a while. There, there's sort of, like, a downside of this, because the model changes so... Well, there's actually, like, there's many kinda downsides that we, that we could talk about. But I think one of them on a personal level is the model changes so often that I sometimes get stuck in this, like, old way of, of thinking about it. And I even find that, like, new people on the team or even new grads that join do stuff in a more kinda [chuckles] like AGI-forward way than I do. So, like, sometimes, for example, I, I, I h- I had this case, like, a couple months ago where there was a memory leak. And so, like, what this is is, you know, like, Claude Code, the memory usage is going up, and at some point it crashes. This is, like, a very common kind of engineering problem that, you know, every engineer has debugged a thousand times. And traditionally, the way that you do it is you take a heap snapshot, you put it into a special debugger, you kinda figure out what's going on, you know, use these special tools to see what's happening. Um, and I was doing this, and I was kinda, like, looking through these traces and trying to figure out what was going on, and the engineer that was newer on the team just, uh, had Claude Code do it. And it was like: "Hey, Claude, it seems like there's a leak. Can you figure it out?" And so, like, Claude Code did exactly the same thing that I was doing. It, it took the heap snapshot. It wrote a little tool for itself, so it can kinda like analyze it itself. Um, it was sort of like a just-in-time program. Uh, and it found the issue and put up a pull request faster than I could. So it's, it's something where, like, for those of us that have been using the model for a long time, you still have to kinda transport yourself to the current moment and not get stuck back in an old model, because it's not Sonnet 3.5 anymore. The new models are just completely, completely different. Uh, and just this, this mindset shift is,

10. 24:02 – 26:48

    Principles for the Claude Code team

    1. BCBoris Cherny24:02

       is very different.

    2. LRLenny Rachitsky24:03

       I hear you have these very specific principles that you've codified for your team, that when people join you, you kinda walk them through them. I believe one of them is, "What's better than doing something? Having Claude do it." And it feels like that's exactly what you described with this memory leak, is just, like, you almost forgot that principle of like, "Okay, let me see if Claude can solve this for me."

    3. BCBoris Cherny24:21

       ... There's this, uh, there's this interesting thing that happens also when you, um, when you underfund everything a little bit, uh, because then people are kind of forced to clodify. And th- this is something that we see. So, you know, for work where sometimes we just put, like, one engineer on a project, and the way that they're able to ship really quickly, because they wanna ship quickly, this is, like, an intrinsic motivation that comes from within. It's just wanting to do a good job. One, if you have a good idea, you just really wanna get it out there. No one has to force you to do that. That comes from you. Um, and, and so i- if you have Claude, you can just use that to automate a lot of work, uh, and that- that's kinda what we see over and over. So I think that's kinda like one principle is underfunding things a little bit. I think another principle is just encouraging people to go faster. So if you can do something today, you should just do it today, and this is something we, we really, really encourage on the team. Early on, it was really important because it was just me, and so our only advantage was speed.

    4. LRLenny Rachitsky25:15

       [chuckles]

    5. BCBoris Cherny25:17

       That's the only way that we could ship a product that would compete in this very crowded coding market. But nowadays, it's still, uh, very much a principle we have on the team, and if you wanna go faster, a really good way to do that is to just have Claude do more stuff. Um, so it, it just very much encourages that.

    6. LRLenny Rachitsky25:32

       This idea of underfunding, it's so interesting because in general, there's this feeling like AI is gonna allow you to not have as many employees, not have as many engineers, and so it's not only you can be more productive. What you're saying is that you will actually do better if you underfund. It's not just that AI can make you faster, it's you will get more out of the AI tooling if you have fewer people working on something.

    7. BCBoris Cherny25:54

       Yeah, if you, if you hire great engineers, they'll figure out how to do it, and, uh, especially if you empower them to do it. This is something I actually talk, talk a lot about with, uh, you know, with, like, CTOs and kinda all sorts of companies. My advice generally is don't try to optimize, don't, don't try to cost cut at the beginning. Start by just giving engineers as many tokens as possible. And now, now you're starting to see companies like, you know, at Anthropic, we have, you know, everyone can use a lot of tokens. We're starting to see this come up as, like, a perk at some companies, where, like, if you join, you get unlimited tokens. This is a thing I very much encourage because, um, it makes people free to try these ideas that would've been too crazy. And then, if there's an idea that works, then you can figure out how to scale it, and that's the point to kind of optimize and to cost cut, figure out, like, you know, maybe you can do it with Haiku or with Sonnet instead of Opus or whatever. But at the beginning, you just wanna throw a lot of tokens at it and see if the idea works, and give engineers

11. 26:48 – 27:55

    Why you should give engineers unlimited tokens

    1. BCBoris Cherny26:48

       the freedom to do that.

    2. LRLenny Rachitsky26:49

       So the advice here is, uh, just be, be loose with your tokens, with the, the cost on, on using these models. People hearing this may be like: Of course, he works at Anthropic. He'd want us to use as many tokens as possible. But you're... What you're saying here is that the most interesting, innovative ideas will come out of someone just kind of taking it to the max and seeing what's possible.

    3. BCBoris Cherny27:08

       Yeah, and I, and I think the reality is, like, at small scale, like, it, you know, you're not gonna get, like, a giant bill or anything like this.

    4. LRLenny Rachitsky27:13

       Mm.

    5. BCBoris Cherny27:13

       Like, if it's an individual engineer e- experimenting, it's still-- the token cost is still probably relatively low relative to their salary or, you know-

    6. LRLenny Rachitsky27:21

       Mm

    7. BCBoris Cherny27:21

       ... other costs of running the business. So it, it's actually, like, not, not a huge cost. As the thing scales up, so, like, let's say, you know, they build something awesome, and then it takes a huge amount of tokens, and then the cost becomes pretty big, that's the point at which you wanna optimize it. But don't, don't do that too early.

    8. LRLenny Rachitsky27:37

       Have you seen companies where their, uh, token cost is higher than their salary? Is that a trend you think we're gonna find and see?

    9. BCBoris Cherny27:44

       You know, at Anthropic, we're starting to see some engineers that are spending, you know, like, hundreds of thousands a month in, in tokens. Um, so we're starting to see this a little bit. Um, there's some companies that are- we're starting

12. 27:55 – 32:15

    Will coding skills still matter in the future?

    1. BCBoris Cherny27:55

       to see similar things. Yeah.

    2. LRLenny Rachitsky27:58

       Going back to coding, do you miss writing code? Is this something you're kind of sad about, that this is no longer a thing you will do as a software engineer?

    3. BCBoris Cherny28:06

       It's funny. For me, uh, you know, like, when, when I learned engineering, for me, it was very practical. Uh, I learned engineering, so I could build stuff, and for me, I was, I was self-taught. You know, like, I studied economics in school, but, um, I didn't study CS. But I, I taught myself engineering kinda ear- early on. I was programming in, like, middle school, and from the very beginning, it was very practical. So I actually, like, I've learned to code so that I can cheat on a math test. That was, like, the first thing [chuckles] we had these, like-

    4. LRLenny Rachitsky28:34

       Wow

    5. BCBoris Cherny28:34

       ... graphing calculators, and the, you know, I just programmed the answer into the-

    6. LRLenny Rachitsky28:37

       TI-83?

    7. BCBoris Cherny28:38

       TI-83 Plus. Yeah, yeah, exactly.

    8. LRLenny Rachitsky28:40

       Plus. [chuckles]

    9. BCBoris Cherny28:41

       Plus, yeah. So I, I programmed the answers in, and then the next, like, math test, whatever, like, the next year, they... It was just, like, too hard. Like, I couldn't program all the answers in, 'cause I didn't know what the questions were, and so I had to write, like, a little solver so that it, it was a program that would just, like, solve these, like, uh, you know, these al- algebra questions or whatever. And then I figured out you can get a little cable, you can give the program to the rest of the class, and then the whole class gets A's. But then we all got caught, and the teacher told us to knock it off.

    10. LRLenny Rachitsky29:07

        [chuckles]

    11. BCBoris Cherny29:07

        But from the very beginning, it's, it's-

    12. LRLenny Rachitsky29:08

        Wow

    13. BCBoris Cherny29:08

        ... always just been very practical for me, where programming is a way to build a thing. It's not the end in itself. At some point, I personally fell into the rabbit hole of kind of like the, the beauty of, of programming. Um, so, like, I, I wrote a book about TypeScript. Um, I sort of, uh, the... Ac- actually, at the time, it was the world's biggest, uh, TypeScript meet-up, just 'cause I fell in love with the language itself. Uh, and I kinda got in deep into, like, functional programming and, and all the stuff. Uh, the, I think a lot of coders, they get distracted by this. For me, it was always sort of, um... They, there is a beauty to programming, and especially to functional programming. There's a beauty to type systems. Um, there, there's a certain kind of, like, this, like, buzz that you get, like, when you solve, like, a really aw- a really complicated, uh, math problem. It's kinda similar when you kinda balance the types or, you know, the program is just, like, really beautiful. But it's really not the end of it. Um, I think for me, coding is very much a tool, and it, it's a way to do things. Uh, that said, not everyone feels this way. So, for example, you know, like, there's one engineer, uh, on the team, Lena, who, you know, is still writing C++ on the weekends by hand because, you know, for her, she just really enjoys writing C++ by hand. And so everyone is different, and I think even as this field changes, even as everything changes-... there's always space to do this. There's always space to enjoy the art, um, and to, and, and to kind of do, do things by hand, uh, if you want.

    14. LRLenny Rachitsky30:34

        Do you worry about your skills atrophying as an engineer? Is that something you worry about, or is it just like, you know, this is just how it's gonna go?

    15. BCBoris Cherny30:41

        I think it's just the way that it, that it happens. I d- I don't worry about it too much, personally. I think, uh, for me, like, programming is on, is on a continuum.

    16. LRLenny Rachitsky30:48

        Mm.

    17. BCBoris Cherny30:48

        And, you know, like, way back in the day, you know, like, software actually is, like, relatively new, right? Like, if you look at the way programs are written today, like, using software that's running on a virtual machine or something, this has been the way that we've been writing programs since probably the nineteen sixties. So, you know, it's been, you know, like, sixty years or something like that. Before that, it was punch cards. Before that, it was switches. Before that, it was hardware, and before that, it was just, you know, like, literally pen and paper. It was like a room, a room full of people that were doing math on, on paper. And so, you know, programming has always changed in this way. In some ways, you still want to understand the layer under the layer because it helps you be a better engineer, and I think this will be the case maybe for the next year or so. Um, but I think pretty soon it just won't really matter. It, it's just gonna be kind of like the, the assembly code writing, running under the program or something like this. Uh, at an emotional level, you know, I, I feel like I've always had to learn new things. And as a programmer, it's actually not-- It doesn't feel that new because there's always new frameworks, there's always new languages. It's just something that we're quite comfortable with in the field. But at the same time, I, you know, this isn't true for everyone, and I think for some people, they're gonna feel a greater sense of, I don't know, maybe like loss or nostalgia or atrophy or something like this.

    18. LRLenny Rachitsky32:01

        I don't know if you saw this, but Elon was saying that, uh, why isn't the AI just writing binary, straight to binary? Uh, because what's the point of all this, you know, programming abstraction in the end?

    19. BCBoris Cherny32:12

        Yeah, it's a good question. I mean, it, it totally can do that if you want

13. 32:15 – 36:01

    The printing press analogy for AI’s impact

    1. BCBoris Cherny32:15

       it to. [chuckles]

    2. LRLenny Rachitsky32:15

       Oh, man. So what I'm hearing here is in term-- There's always this question, should I learn to code? Should people in school learn to code? W- uh, what I heard from you is your take is in, like, a year or two, you don't really need to.

    3. BCBoris Cherny32:27

       My take is, I think for, for people that are using, um, that are, that are using Claude Code, that are using agents to code today, you still have to understand the layer under. But yeah, in a year or two, it's not gonna matter. I w- I was thinking about, um, what is the right, like, historical analog for this? 'Cause, y- like, like, somehow we have to situate this thing in history and, and kind of figure out when have we gone through similar transitions? What's the right kinda mental model for this? I think the thing that's come closest for me is the printing press. And so, you know, i- if you look at Europe in, uh, you know, like, in the, in the, in the mid, the mid-fourteen hundreds, literacy was actually very low. Uh, there was sub one percent of the population, it was scribes, that, uh, you know, they were the ones that did all the writing. They, they were the ones that did all the reading. They were employed by, like, lords and kings that often were not literate themselves. And so, you know, it was their job of this very tiny percent of the population to do this. And at some point, there, you know, Gutenberg and the, and the printing press came along, and there was this crazy stat that in the fifty years after the printing press was, uh, built, there was more printed material created than in the sen- in the, in the thousand years before. And so the, the volume of printed material just went way up. Uh, the cost went way down. It went down something like a hundred X over the next fifty years. And if you look at literacy, you know, oh, it, it actually took a while, because learning to read and write is, you know, it's quite hard. It takes an education system. It takes free time. You-- It takes, like, not having to work on a farm all day, so that you actually have time for education and things like this. But over the next two hundred years, it went up to, like, seventy percent globally. So I think this is the kind of thing that we might see is a similar kind of transition. And there was, uh, there was actually this interesting, um, historical document where there, there was an interview with some, like, scribe in the fourteen hundreds about, like, how do you feel a- about the printing press? And they were actually very excited because they were like: "Actually, the thing that I don't like doing is copying between books. The thing that I do like doing is drawing the art in books and then doing the bookbinding, and I'm really glad that now my time is freed up." And it, it's interesting, like, as an engineer, I sort of felt, like, a peril with this. Like, this is sort of how I feel, where I don't have to do the tedious work anymore of coding, because this has always been sort of the detail of it. It's always been the tedious part of it, and kind of like messing with a git and kinda using all these different tools. That, that was not the fun part. The fun part is figuring out what to build and k- coming up with this. It's, uh, it's talking to users. It's thinking about these big systems. It's, it's thinking about the future. It's collaborating with, you know, other people on the team, and that, that's what I get to do more of now.

    4. LRLenny Rachitsky35:07

       And what's amazing is that the tool you're building allows anybody to do this. People that have no technical experience can do exactly what you're describing. Like, I've, I've been doing a bunch of random little projects, and at any-- it's just like, anytime you get stuck, just like: "Help me figure this out," and you get unblocked. Like, I used to... I, I was an engineer for, in earlier in my career, for ten years, and I just remember spending so much time on, like, libraries and dependencies and little things and just like: "Oh, my God, what do I do?" And then looking on Stack Overflow, and now it's just like: "Help me figure this out, and here's step by, step one, two, three, four. Okay, we got this."

    5. BCBoris Cherny35:39

       Yeah, exactly, exactly. I was talking to an engineer earlier today. They're like, they're writing some service in Go, and, you know, it's been, like, a month already, and they, they built up the service. Like, it's, it's working quite well. And then I was like: "Okay, so, like, how do you feel writing?" And he was like: "You know, like, I, I still don't really know Go, but..." [chuckles]

    6. LRLenny Rachitsky35:54

       [chuckles]

    7. BCBoris Cherny35:55

       And I, I think we're gonna start to see more and more of this. It's like, if you know that it works correctly and efficiently, then you, you don't actually

14. 36:01 – 40:41

    Which roles will AI transform next?

    1. BCBoris Cherny36:01

       have to know all the details.

    2. LRLenny Rachitsky36:02

       Clearly, the life of a software engineer has changed dramatically. It's like a whole new job now, as of the past year or two. What do you think is the next role that will be most impacted by AI within, either within tech, like, you know, product managers, designers, or even outside tech? Just like, what do you think? Where do you think AI is going next?

    3. BCBoris Cherny36:23

       I think it's gonna be a lot of the roles that are adjacent to engineering. Um, so yeah, it could be, like, product managers, it could be design, could be data science.... it is gonna expand to pretty much any kind of work that you can do on a computer because the model is just gonna get better and better at this. Um, and, you know, like, this is the Cowork product is kind of the first way to get at this, but it's just the first one. And it's the thing that I think brings AI to-- a- agentic AI to people that haven't really used it before, and people are starting just to, to, to get a sense of it for the first time. When I think back to engineering a year ago, no one really knew what an agent was, no one really used it, but nowadays, it's just the way that, you know, we do, we do our work. And then when I look at non-technical work today, um, so, you know, like, you know, or like maybe semi-technical, like product work and, you know, like data science and things like this, when you look at the kinds of AI that people are using, it's al-- it's always these, like, conversational AI. It's like a chatbot or whatever, but no one really has used an agent before. And this word agent just gets thrown around all the time, and it's just, like, so misused. It's, like, lost all meaning. But agent actually has, like, a very specific technical meaning, which is it's a, it's a AI, it's a LLM that's able to use tools. So it doesn't just talk, it can actually act, and it can interact with your system. And, you know, this means, like, it can use your Google Docs, and it can, it can send email. It can run commands on your computer and do all this kind of stuff. So I think, like any kind of job where you do-- you use computer tools in this way, I think this is gonna be next. This is something we have to kinda figure out as a, as a society. This is something we have to figure out as an industry. Um, and I think for me also, this is one of the reasons it, it, it feels very important and urgent to do this work at Anthropic, because I think we take this very, very seriously. Um, and so now, you know, we have economists, we have, uh, policy folks, we have social impact folks. This is something we just wanna talk about a lot, so as a society, we can kinda figure out what to do, 'cause it shouldn't be up to us.

    4. LRLenny Rachitsky38:19

       So the big question, and which is you're kind of alluding to, is jobs and job loss and things like that. There's this concept of Jevons Paradox, of just as we can do more, we hire more, and it's not actually as scary as it looks. What have you experienced so far, I guess, with AI becoming a big part of the engineering job? Just are you hiring more than if you didn't have AI? And just thoughts on jobs.

    5. BCBoris Cherny38:41

       Yeah, I mean, for our team, we're, we're hiring. Um, so Claude Code team is hiring. Um, if you're interested, just check out the jobs page on, on Anthropic. Personally, it's, you know, all, all this stuff has just made me enjoy my work more. I have never enjoyed coding as much as I do today because I don't have to deal with all the minutiae. So for me, personally, it's been quite exciting. This is something that we hear from a lot of customers, where they love the tool, they love Claude Code because it just makes coding delightful again. Uh, and that's just, that's just so fun for them. But it's hard to know where this thing is gonna go. And a- again, I just, like, I have to reach for these historical analogues. Uh, and I, I think the printing press is just such a good one because what happened is this technology that was locked away to a small set of people, like knowing how to read and write, became accessible to everyone. It was just inherently democratizing. Everyone started to be able to do this. And if that wasn't the case, then something like the Renaissance just could never have happened. Because a lot of the Renaissance, it was about, like, knowledge spreading. It was about, like, written records that people used to communicate, um, you know, 'cause there, there were no phones or anything like this. There was, there was no internet at the time. So it's, it's about like, what does this enable next? And I think that's the very optimistic version of it for me, and that's the part that I'm really excited about. It's just unimaginable. You know, like, we couldn't be talking today if the printing press hadn't been invented. Like, our microphones wouldn't exist. None of the things around us would exist. It just wouldn't be possible to coordinate such a large group of people if that wasn't the case. And so I imagine a world, you know, a few years in the future where everyone is able to program, and what does that unlock? Anyone can just build software anytime, and I have no idea. It's just the same way that, you know, in the 1400s, no one could have predicted this. Um, I think it's the same way, but I do think in the meantime, it's gonna be very disruptive, and it's gonna be painful for a lot of people. Um, and i- a- again, as a society, this is a conversation that we have to have, and this is a thing that we have

15. 40:41 – 44:37

    Tips for succeeding in the AI era

    1. BCBoris Cherny40:41

       to figure out together.

    2. LRLenny Rachitsky40:42

       So for folks hearing this that want to succeed and, you know, make it in this crazy turmoil we're entering, any advice? Is it, you know, play with AI tools, get really proficient at the latest stuff? Is there anything else that you recommend to help people, uh, stay ahead?

    3. BCBoris Cherny40:58

       Yeah, I think that's pretty much it. Uh, experiment with the tools, get to know them. Don't be scared of them. Um, just, you know, dive in, try them. Be on the bleeding edge, be on the frontier. Maybe the second piece of advice is try to be a generalist more than you have in the past. For example, in school, a lot of people that study CS, they learn to code, and they don't really learn much else. Maybe they learn a little bit of systems architecture or something like this. But some of the most effective engineers that I work with every day, and some of the most effective, you know, like, product managers and so on, they cross over disciplines. So on the Claude Code team, everyone codes. You know, our product manager codes, our engineering manager codes, our designer codes, our finance guy codes, our data scientist codes. Like, everyone on the team codes. And then, and then if I look at particular engineers, people often cross different disciplines. So some of the strongest engineers are hybrid product and infrastructure engineers, or product engineers with really great design sense, and they're able to do design also. Or an engineer that has a really good sense of the business and can use that to figure out what to do next. Or an engineer that also loves talking to users and can just really channel what, what users want, uh, to figure out what's next. So I think a lot of the people that will be rewarded the most over the next few years, they won't just be AI native, and they don't just know how to use these tools really well, but also, they're curious, and they're generalists, and they cross over multiple disciplines and can think about the broader problem they're solving, rather than just the engineering part of it.

    4. LRLenny Rachitsky42:29

       Do you find these three separate disciplines still useful as a way to think about the team? They're, you know, engineering, design, uh, product management. Do you find, like, those, even though they are now coding and contributing to-... thinking about what to build, do you feel like those are three roles that will persist long-term, at least at this point?

    5. BCBoris Cherny42:46

       I think in the short term it'll persist, but one thing that we're starting to see is there's maybe a fifty percent overlap in these roles, where a lot of people are actually just doing the same thing, and some people have specialties. For example, I code a little bit more versus Kat, our PM, does a little bit more, you know, coordination or planning or, you know, forecasting or things like that.

    6. LRLenny Rachitsky43:04

       Stakeholder alignment.

    7. BCBoris Cherny43:06

       Stakeholder alignment, exactly. I, I do think that there is a future where I think by the end of the year, what we're gonna start to see is these start to get even murkier, murkier, where i, I think in some places, the title software engineer is gonna start to go away and it's just gonna be replaced by builder, or maybe it's just everyone's gonna be a product manager and everyone codes or something like this.

    8. LRLenny Rachitsky43:26

       Who says hiring has to be fair? Every founder and hiring manager I've been speaking with these days is feeling the same pressure: hire the best people as fast as possible. But recruiting is time-consuming, alignment is hard, and competition for great talent keeps getting tighter. That's why teams like Eleven Labs, Brex, Replit, Deal, and five thousand other organizations use Metaview, the AI company giving high-performance teams a real unfair advantage in hiring. They give you a suite of AI agents that behave like recruiting coworkers. They find candidates for you based on your exact criteria, take interview notes automatically, gather insights across your hiring process, and help you identify the best candidates in your pipeline. AI handles the recruiting toil and gives you a real source of truth. That means hours saved per hire and a team focused on what matters most, winning the right candidates. Don't let your competitors outhire you. Metaview customers close roles thirty percent faster. Try Metaview today for free and get an extra month of sourcing at metaview.ai/lenny. That's M-E-T-A view.ai/lenny.

16. 44:37 – 46:32

    Poll: Which roles are enjoying their jobs more with AI

    1. LRLenny Rachitsky44:38

       You talked about how you're enjoying coding more. I actually did this little informal survey on Twitter, I don't know if you saw this, where I just asked-- I did three different polls. I asked engineers, "Are you enjoying your job more or less since adopting AI tools?" And then I did a separate one for PMs and one for designers. And both engineers and PMs, seventy percent of people said they are enjoying their job more, and about ten percent said they're enjoying their job less. Designers, interestingly, only fifty-five percent said they are enjoying their job more, and twenty percent said they're enjoying their job less. I thought that was really interesting.

    2. BCBoris Cherny45:11

       That's super interesting. I'd, I'd love to talk to these people, uh, you know, both in the more bucket and the less bucket, just to understand. Did, did you get to follow up with any of them?

    3. LRLenny Rachitsky45:20

       They... A few people replied, and we're actually doing a follow-up poll that we'll link to in the show notes of going deeper into some of this stuff. But a lot of-- there's, like, you know, the factors that make it more fun and less fun. Uh, the designers, they didn't share a lot, actually, of just, like, the people that are... I actually asked, just like: "Why are you enjoying your job less?" I didn't hear a lot, so I'm curious what's going on there.

    4. BCBoris Cherny45:37

       Yeah, I, I'm seeing this a little bit with, uh, at, at Anthropic, I think everyone is fairly technical. This is something that we screen for, you know, when, when people join. We have, uh-- there, there's a lot of technical interviews that people go, go through, even for non-technical functions. Uh, and, you know, our designers virtually code. So I think for them, this is something that they have enjoyed from what I've seen, because now, instead of bugging engineers, they can just, like, go in and code. And even some designers that didn't code before have just started to do it, and for them, it's great 'cause they can unblock themselves. But I'd be really interested just to hear more people's experiences, 'cause I, I, I bet it's not uniform like that.

    5. LRLenny Rachitsky46:18

       Yeah, so maybe if you're listening to this, leave a comment if you're finding your job's less fun and you're enjoying your job less. 'Cause what you're saying and what I'm hearing from most people, seventy percent of PMs and engineers are loving their job more. That's... Like, if you're not in that bucket, you could-- something's going on.

17. 46:32 – 51:53

    The principle of latent demand in product development

    1. BCBoris Cherny46:32

       Yeah, yeah. Uh, we, we do see that people use also different tools. So, for example, our designers, they use, uh, the Claude desktop app a lot more to, to do their coding. So you just download the desktop app, there's a code tab, uh, it's right next to Cowork. And it's actually the same as that Claude Code, so it's, like, the same agent and everything. We've had this for, you know, for many, many months. Uh, and so you can use this to code in a way that you don't have to open a bunch of terminals, but you still get the power of Claude Code, and the biggest thing is you can just run as many, you know, Claude sessions in parallel as you want. We can-- you know, we call this multi-Clauding.

    2. LRLenny Rachitsky47:04

       Mm.

    3. BCBoris Cherny47:04

       So this is a... It's a, it's a little more native, I think, for folks that are not engineers, and really this is back to bringing the product to where the people are. You don't wanna make people use a different workflow. You don't wanna make them go out of their way to learn a new thing. It's whatever people are doing, if you can make that a little bit easier, then that's just gonna be a much better product that people enjoy more. And this is just this principle of latent demand, which I, I think is just the, the single most important principle in product.

    4. LRLenny Rachitsky47:29

       Can you talk about that, actually? 'Cause I was gonna go there. Explain what this principle is and, and, and just what happens when you u- unlock this latent demand.

    5. BCBoris Cherny47:37

       Latent demand is this idea that if you build a product in a way that can be hacked or can be kinda mi- misused by people in a way it wasn't really designed for, to do kinda something that they wanna do, then this helps you, as the product builder, learn where to take the product next. So an example of this is, uh, Facebook Marketplace. So the, the manager for the team, Fiona, she, she was actually the founding manager for, uh, the marketplace team, and she talks about this a lot. Facebook Marketplace is sort of based on the observation back in, uh, this must have been, like, twenty, twenty sixteen or so- or something like this, that forty percent of posts in Facebook groups are buying and selling stuff. So this is crazy. It's like people are, are abusing the Facebook Groups product to buy and sell, and it's not, it's not abuse in kind of like a security sense. It's abuse in that no one designed the product for this, but they're kind of figuring it out because it's, it's just so useful for this. And so it was pretty obvious, if you build a better product to let people buy and sell, they're gonna like it. And it was just very obvious that Marketplace would be a hit from this. And so the first thing was buy and sell groups, so kind of special purpose groups to let people do that, and the second product was Marketplace.... Ah, Facebook Dating, I think, started in a pretty similar place. And I think that w- the observation was, if you look at people looking at-- if you look at, ah, profile views, so people looking at each other's profiles on Facebook, sixty percent of profile views were people that are not friends with each other, that are opposite gender. And so this is this kind of like, you know, like, traditional kinda date- dating setup, uh, you know, people are just, like, creeping on each other. So maybe if you can build a product for this, it's, you know, it, it might work. Um, and so i- th- this idea of latent demand, I, I think is just so powerful in that... For example, this is also where Cowork came from. We saw that for the last six months or so, a lot of people using Claude Code were not using it to code. There was someone on Twitter that was using it to grow tomato plants. There was someone else using it to analyze their genome. Someone was using it to, uh, recover photos from a corrupted hard drive. It was, like, c- uh, wedding photos. Uh, there was someone that was using it for, uh, I think, like, uh, they, they, they were using it to analyze a MRI. So there, there's just all these different use cases that are not technical at all, and it was just really obvious. Like, people are jumping through hoops to use a terminal to do this thing. Maybe we should just build a product for them. And we saw this actually pretty early. Back in maybe May of last year, I remember walking into the office, and our data scientist, Brendan, was- had a Claude Code on his, uh, computer. He just had a terminal up, and I was like-- I was shocked. I was like: "Brendan, what are, what are you doing? Like, you, you figured out how to open the terminal," which is, you know, it's a, it's a very engineering product. Even a lot of engineers don't wanna use a terminal. It's just like a... It's like, just like the lowest-level way to, to do your work. Um, just really, really, uh, kinda in the weeds of the computer. And so he figured out how to use the terminal. He downloaded Node.js, he downloaded Claude Code, and he was doing SQL analysis in the terminal, and it wa- it was crazy. And then the next week, all the data scientists were doing the same thing. So when you see people abusing the product in this way, using it in a way that it wasn't designed in order to do something that is useful for them, it's just such a strong indicator that you should just build a product and, and people are gonna like that. It's something that's special purpose for that. I think now there, there's also this kinda interesting second dimension to latent demand. This is sort of the traditional framing is: Look at what people are doing, make that a little bit easier, empower them. The modern framing that I've been seeing in the last six months is a little bit different, and it's look at what the model is trying to do and make that a little bit easier. And so wh- when we first started building Claude Code, I think a lot of the way that people approached designing things with LLMs is they kind of put the model in a box, and they were, "Here's this application that I wanna build. Here's the thing that I want it to do. Model, you're gonna do this one component of it. Here's the way that you're gonna interact with these tools and APIs and whatever." And for Claude Code, we inverted that. We said: The product is the model. We wanna expose it. We wanna put the minimal scaffolding around it, give it the minimal set of tools, so it can do the things. It can decide which tools to run. It can decide in what order to run them in, and so on. And I, I think a lot of this was just based on kind of latent demand of what the model wanted to do. And so in research, we call this being on distribution. Uh, you wanna see, like, what the model is trying to do. In product terms, latent demand is just the same exact

18. 51:53 – 54:04

    How Cowork was built in just 10 days

    1. BCBoris Cherny51:53

       concept but applied to a model.

    2. LRLenny Rachitsky51:55

       You talked about Cowork. Something that I saw you talk about when you launched that initially is you-- your team built that in ten days. That's insane.

    3. BCBoris Cherny52:02

       Yeah.

    4. LRLenny Rachitsky52:02

       Uh, [chuckles] it came out, I think it was like, you know, used by millions of people pretty quickly, something like that being built in ten days. Uh, anything there, any stories there, other than just, it was just, you know, we used Claude Code to build it, and that's it?

    5. BCBoris Cherny52:14

       Yeah, it, it, it's funny, uh, Claude Code, like I said, when we released it, it was not immediately a hit. It became a hit over time, and there was a few inflection points. So one was, you know, like Opus-four, uh, it just really, really inflected, and then in November, it inflected, and it just keeps inflecting. It- the growth just keeps getting steeper and steeper and steeper every day. But, you know, for the first few months, it wasn't a hit. Uh, people used it, but a lot of people couldn't figure out how to use it. They didn't know what it was for. The model still, like, wasn't very good. Cowork, when we released it, it was just immediately a hit, much more so than Claude Code was early on. I think a lot of the credit, honestly, just goes to, like, Felix and, and Sam and the-- and Jenny and the, the team that built this. It's just an incredibly strong team. And again, the, the place Cowork came from is just this latent demand. Like, we saw people using Claude Code for these non-technical things, and we're trying to figure out, what do we do? And so for a few months, the team was exploring. They were trying all sorts of different options, and in the end, someone was just like: "Okay, what, what if we just take Claude Code and put it in the desktop app?" And that's essentially the thing that worked. And so over ten days, they just completely used Claude Code to build it. Uh, and, you know, Cowork is actually... The- there's this very sophisticated security system that's, that's built in, and essentially, these guardrails to make sure that the model kinda does the right thing, it doesn't go off the rails. So, for example, we ship an entire virtual machine with it, and Claude Code just wrote all of this code. So we just had to think about, all right, how do we make this a little bit safer, a little more self-guided for, uh, people that are not engineers? It was fully implemented with Claude Code, took about ten days. We launched it early. You know, it was still pretty rough, and it's still pretty rough around the edges, but this is kind of the way that we learn, um, both on the product side and on the safety side, is we have to release things a little bit earlier than we think so that we can get the feedback, so that we can talk to users. We can understand what people want, and then that will shape where

19. 54:04 – 59:35

    The three layers of AI safety at Anthropic

    1. BCBoris Cherny54:04

       the product goes in the future.

    2. LRLenny Rachitsky54:05

       Yeah, I think that point is so interesting, and, and it's so unique. It- there's always been this idea, release early, learn from users, get feedback, iterate. The fact that it's hard to even know what the AI is capable of and how people will try to use it is, like, is a ne- unique reason to start releasing things early. So that'll help you, as you s- exactly describe this idea of: What is the latent demand in this thing that we didn't really know? Let's put it out there and see what people do with it.

    3. BCBoris Cherny54:30

       Yeah, and, and for Anthropic, as a safety lab, the other dimension of that is safety, 'cause, um, you know, like, w- when you think about model safety, there's a bunch of different ways to study it. Sort of the lowest level is alignment and mechanistic interpretability. So this is when we train the model, we wanna make sure that it's safe. We-... at this point, have like pretty sophisticated technology to understand what's happening in the neurons to trace it. And so, for example, like, if there's a neuron related to deception, we can start-- we're, we're starting to get to the point where we can monitor it and understand that it's activating. Um, and so this is just, this is alignment, this is mechanistic interpretability. It's, like, the lowest layer. The second layer is evals, and this is essentially a laboratory setting. The model is in a petri dish, and you study it, and you put it in a synthetic situation and just say: "Okay, like, model, what do you do? And are you doing the right thing? Is it aligned? Is it safe?" And then the third layer is seeing how the model behaves in the wild. And as the model gets more sophisticated, this be- this becomes so important because it might look very good on these first two layers, but not great on the third one. We released Claude Code really early because we wanted to study safety, and we actually used it within Anthropic for, I think, four or five months or something before we released it because we weren't really sure. Like, this is the first agent that, you know, the first big agent that I think folks had released at that point. Um, it was definitely the first, uh, you know, coding agent that became broadly used, and so we weren't sure if it was safe. And so we actually had to study it internally for a long time before we felt good about that. And even since, you know, there's a lot that we've learned about alignment, there's a lot that we've learned about safety that we've been able to put back into the model, back into the product. And for Cowork, it's pretty similar. Uh, the model's in this new setting. It's, you know, doing these tasks that are not engineering tasks. It's an agent that's acting on your behalf. It looks good on alignment. It looks good on evals. We tried it internally, it looks good. We tried it with a few customers, it looks good. Now we have to make sure it's safe in the real world. And so that's why we released a little early. That's why we call it a research preview. Um, but yeah, it's just-- it's constantly improving, um, and this is really the only way to, to make sure that over the long term, the model is aligned and it's doing the right things.

    4. LRLenny Rachitsky56:33

       It's such a wild space that you work in, where there's this insane competition and pace. At the same time, there's this fear that if you get some-- if the, the, you know, the god can escape and cause damage, and just finding that balance must be so challenging. What I'm hearing is there's kind of these three layers, and I know there's like... This could be a whole podcast conversation, is how you all think about [chuckles] the safety piece, but just what I'm hearing is there's these three layers you work with. Uh, there's kind of like observing the model, thinking and operating. There's e-- tests, evals that tell you, "Is this doing bad things?" And then releasing it early. I haven't actually heard a ton about that first piece. That is so cool. So you guys can... Y- there's an observability tool that can let you peek inside the model's brain and see how it's thinking and where it's heading?

    5. BCBoris Cherny57:16

       Yeah, you should, uh, you should at some point have Chris Olah on the podcast because, uh, he, he's just the industry expert on this. He, he invented this field of, uh, we call it mechanistic interpretability. Uh, and the, the idea is, uh, you know, like at, at its core, like, what is your brain? Like, what are... W- what is it? It's like a, it's a bunch of neurons that are connected, and so what you can do is, like, in a human brain or a, a animal brain, you can study it at this kind of mechanistic level to understand what the neurons are doing. It turns out, surprisingly, a lot of this does translate to models also. So model neurons are not the same as animal neurons, but they behave similarly in a lot of ways, and so we've been able to learn just a ton about the way these neurons work, about, you know, this layer or this neuron maps to this concept, how particular concepts are encoded, how the model does planning, how it, how it thinks ahead. You know, like a, a long time ago, we weren't sure if the model was just predicting the next token or is doing something a little bit deeper. Now, I think there's actually quite strong evidence that it is doing something a little bit deeper, and then the structures the way to do this are pretty sophisticated now, where as the models get bigger, it's not just like a single neuron that corresponds to a concept. A single neuron might correspond to a dozen concepts, and if it's activated together with other neurons, this is called superposition, and, uh, together it represents this more sophisticated concept. And it's just something we're learning about all the time. You know, and, uh, for Anthropic, as, as we think about the way this space evolves, doing this in a way that is safe and good for the world is just... This is the reason that we exist, and this is the reason that everyone is at Anthropic. Uh, everyone that is here, this is the reason why they're here. So a lot of this work, we actually open source, uh, we publish it a lot, um, and, you know, we publish very freely to talk about this, just so we can inspire other labs that are working on similar things to do it in a way that's safe. And this is something that we've been doing for Claude Code also. We call this the race to the top, uh, internally. And so for Claude Code, for example, we released an open-source sandbox, and this is a sandbox that you can run the, the agent in, and it just makes sure that there are certain boundaries, and it can't access like, everything on your system. Uh, and we made that open source, and it actually works with any agent, not just Claude Code, because we wanted to make it really easy for others to do the same thing. Um, so this is just the same principle, a race to the top. Um, we, we wanna make sure this thing goes well, and this is just the,

20. 59:35 – 1:02:25

    Anxiety when AI agents aren’t working

    1. BCBoris Cherny59:35

       this is the lever that we have.

    2. LRLenny Rachitsky59:37

       Incredible. Okay, I definitely wanna spend more time on that. I, I will follow up with this suggestion. Something else that I've been noticing in the, in the field across engineers and product managers, others that work with agents, is there's this kind of anxiety people feel when their agents aren't working. There's a sense that like, "Oh, man, uh, Niza has a question I need to answer," or, "It's, like, blocked on something," or it's... Or, uh, it's just like, I- I'm like, "There's all this productivity I'm losing. I can't, like, I need to wake up and get it going again." Is that something you feel? Is that something your team feels? Do you feel like this is a, a problem we need to track and think about?

    3. BCBoris Cherny1:00:11

       I always have a bunch of agents running, so, like, at the moment, I have, like, five agents running. And a- at any moment, like, you know, like I, I wake up, and I, I start a bunch of agents. Like, the first thing I did when I woke up was like: "Oh, man, I, I want, I really want to check this thing." So, like, I opened up my phone, Claude iOS app, code tab, uh, you know, like, agent, do, do blah, blah, blah. 'Cause I, I wrote some code yesterday, and I was like: "Wait, did, did I do this right?" I was, like, kinda double, double-guessing something, and it, and it was correct. But now it's just, like, so easy to do this. So I don't know. There, there is this little bit of anxiety maybe. I personally haven't really felt it just 'cause I have agents running all the time, um, and I'm also just, like, not logged into a terminal anymore. Maybe a third of my code now is in the terminal, but also a third is, uh, using the desktop app.... And then a third is the iOS app, which is just so surprising, 'cause I did not think that this would, would be the way that I code, uh, in, even in twenty twenty-six.

    4. LRLenny Rachitsky1:01:03

       I love that you just describe it as coding still, which is just talking to the [chuckles] to Claude Code to code for you, essentially. And it's interesting that this is now, like, this is now coding. Coding now is describing what you want, not writing actual code.

    5. BCBoris Cherny1:01:16

       I, I, I kind of wonder if, uh, the people that used to code using punch cards or whatever, if you show them software-

    6. LRLenny Rachitsky1:01:22

       Mm-hmm

    7. BCBoris Cherny1:01:22

       -what they would have said.

    8. LRLenny Rachitsky1:01:22

       Isn't that crazy? [chuckles]

    9. BCBoris Cherny1:01:23

       And I, I remember reading something, this was maybe, like, very early versions of, like, ACM, uh, like, like, magazine or something, where people were saying, "No, it's not the same thing. Like, this isn't, this isn't really coding." Uh, and, you know, like they, they called it programming. I think coding is kind of a new word.

    10. LRLenny Rachitsky1:01:38

        Mm.

    11. BCBoris Cherny1:01:39

        But I kinda think about this, like in the, back in the... You know, my family is from the Soviet Union. I w-- you know, I, I was born in Ukraine, um, and my grandpa was actually one of the first programmers in the Soviet Union, and he programmed using punch cards. And, uh, you know, like, he, he told-- My mom, uh, growing up told these stories of like, or she, she told these stories of when she was growing up, he would bring these punch cards home, and th- there was th- these, like, big stacks of punch cards. And for her, she would, like, draw all over them with crayons, and that was, like, her childhood memory. But for him, that was, like, his experience of programming, and he actually never saw the software transition. But at some point, it did transition to software, and I think there was probably this older generation of programmers that just didn't take software very seriously, and they, they would have been like, "Well, you know, it's not really coding." But I, I think this is a field that just has always

21. 1:02:25 – 1:03:21

    Boris’s Ukrainian roots

    1. BCBoris Cherny1:02:25

       been changing in this way.

    2. LRLenny Rachitsky1:02:27

       Uh, I don't think you know this, but I was born in Ukraine also.

    3. BCBoris Cherny1:02:30

       Oh, I don't know that.

    4. LRLenny Rachitsky1:02:31

       Yeah.

    5. BCBoris Cherny1:02:31

       Yeah. Which town?

    6. LRLenny Rachitsky1:02:32

       I'm from, I'm from Odessa.

    7. BCBoris Cherny1:02:34

       Oh, me too!

    8. LRLenny Rachitsky1:02:36

       What? [chuckles]

    9. BCBoris Cherny1:02:36

       Yeah. That's crazy. [chuckles]

    10. LRLenny Rachitsky1:02:39

        Wow. Incredible. What a moment. Uh, maybe we're related in some small way.

    11. BCBoris Cherny1:02:44

        Yeah.

    12. LRLenny Rachitsky1:02:44

        Uh, what year did your-- did you leave and your family leave?

    13. BCBoris Cherny1:02:48

        Uh, we came in ninety-five.

    14. LRLenny Rachitsky1:02:50

        Okay. We left in eighty-eight, a little earlier.

    15. BCBoris Cherny1:02:52

        Oh, yeah.

    16. LRLenny Rachitsky1:02:53

        What a different life that would have been to not, to not leave, huh?

    17. BCBoris Cherny1:02:57

        Yeah, I just... I feel, I feel so lucky every day that I get, I get to grow up here.

    18. LRLenny Rachitsky1:03:02

        Yeah. My family, anytime there's, like, a toast or a meal, they're just like, "To America!"

    19. BCBoris Cherny1:03:06

        Yeah. [chuckles]

    20. LRLenny Rachitsky1:03:07

        You know, it's like, okay, enough about that. But you get it, you know, once you start really thinking about what life could have been.

    21. BCBoris Cherny1:03:12

        Yeah, yeah, exactly.

    22. LRLenny Rachitsky1:03:13

        Yeah.

    23. BCBoris Cherny1:03:13

        Yeah, we do the, we do the same toast, but it's still vodka.

    24. LRLenny Rachitsky1:03:15

        [chuckles] It's still vodka. Absolutely. [chuckles] Oh, man. Okay, let me ask you

22. 1:03:21 – 1:08:38

    Advice for building AI products

    1. LRLenny Rachitsky1:03:21

       a couple more things here. You shared some really cool tips for how to get the most out of AI, how to build on AI, how to build great products on AI. One tip you shared is, give your team as many tokens as they want, just, like, let them experiment. You also shared just advice generally of just build towards the model, where the model is going, not to where it is today. What other advice do you have for folks that are trying to build AI products?

    2. BCBoris Cherny1:03:43

       I'd probably share a few more things. So one is, don't try to box the model in. Um, I, I think a lot of people's instinct when they build on the model is they try to make it behave a very particular way. They're like, "You know, this is a component of a bigger system." I, I think some examples of this are people layering, like, very strict workflows on the model, for example. You know, to say, like, "You must do step one, then step two, then step three," and you have this, like, very fancy orchestrator doing this. But actually, almost always, you get better results if you just give the model tools, you give it a goal, and you let it figure it out. I think a year ago, you actually needed a lot of the scaffolding, but nowadays, you don't really need it. So, you know, I, I don't know what to call this principle, but it's like, you know, like, ask not what the model can do for you. Maybe, maybe it's something like this. Just think about, how do you give the model the tools to do things? Don't try to over-curate it. Don't try to put it into a box. Don't try to give it a bunch of context up front. Give it a tool so that it can get the context it needs. You're just gonna get better results. I think a second one is, um, maybe actually, like, a, a more, even more general version of this principle is just the bitter lesson. Uh, and a- actually, for the Claude Code team, we have a... You know, hopefully, hopefully, um, listeners have, have read this, but Rich Sutton had this blog post maybe ten years ago called The Bitter Lesson, uh, and it's actually a really simple idea. His idea was that the more general model will always outperform the more specific model. And I think for him, he was talking about, like, self-driving cars and other domains like this. But actually, there's just so many corollaries to the bitter lesson, and for me, the biggest one is just always bet on the more general model. And, uh, you know, over the long term, like, don't, don't try to use tiny models for stuff. Don't try to, like, fine-tune. Don't try to do any of this stuff. There's, like, some applications, you know, there's some reasons to do this, but almost always try to bet on the more general model if you can, if you have that flexibility. Um, and so these workflows are essentially a way that, uh, you know, it's, it's not, it's not a general model. It's putting the scaffolding around it. And in general, what we see is maybe scaffolding can improve performance maybe ten, twenty percent, something like this, but often these gains just get wiped out with the next model. Uh, so it's almost better to just wait for the next one. And I think maybe this is a final principle and something that Claude Code, I think, got right in hindsight. From the very beginning, we bet on building for the model six months from now, not for the model of today. And for the very early versions of the product, I just wrote so little of my code, 'cause I, I didn't trust it. 'Cause, you know, it was like Sonnet three point five, then it was like three point six, or forget, three, three point five new, whatever, [chuckles] whatever, whatever name we gave it. Um, these models just weren't very good at coding yet. Um, they were, they were getting there, but it was still pretty early. So back then, the model did, uh, you, you used Get For Me. It automated some things, but it, it really wasn't doing a huge amount of my coding. And so the bet with Claude Code was, at some point, the model gets good enough that it can just write a lot of the code. And this is a thing that we first started seeing with Opus4 and Sonnet4, and Opus4 was our first kinda ASL3 class model, uh, that we released back in May. And we just saw this inflection, because everyone started to use Claude Code for the first time, and that, that was kind of when our growth really went exponential. And like I said, it's kinda... It, it stayed there.... So I think this is some-- this is advice that I actually give to, to a lot of folks, especially people building startups. It's gonna be uncomfortable 'cause your product market fit won't be very good for the first six months. But if you build for the model six months out, when that model comes out, you're just gonna hit the ground running, and the product is gonna click and, and start to work.

    3. LRLenny Rachitsky1:07:15

       And when you say build for the model six months out, what is, what is it that you think people can assume will happen? Is it just generally it will get better at things? Is it just like, okay, it's, like, almost good enough, and that's a sign that it'll probably get better at that thing? Is there any advice there?

    4. BCBoris Cherny1:07:30

       I think that's a good way to do it. Like, uh, you know, obviously, within a AI lab, we get to see the specific ways that it gets better.

    5. LRLenny Rachitsky1:07:36

       [chuckles]

    6. BCBoris Cherny1:07:37

       So it's a, it's a little unfair, but we, we also-- we try to talk about this. So, you know, like, one of the ways that it's gonna get better is it's gonna get better and better at using tools and using computers. This is a bet that I would make. Uh, another one is it's gonna get better and better for long-- for running, uh, for long periods of time. And this is a place, you know, like, there's all sorts of studies about this, but if you just trace the tr- trajectory or, you know, maybe even, like, from my own experience, when I used Sonnet 3.5 back, you know, a year ago, it could run for maybe fifteen or thirty seconds before [chuckles] before it started going off the rails, and you just really had to hold its hand through any kind of complicated task. But nowadays, with Opus 4.6, you know, on average, it'll run maybe ten, thirty, twenty, thirty minutes unattended, and I'll just, like, start another Claude and have it do something else. And, you know, like I said, I always have a bunch of Claudes running. Uh, and they can also run for hours or even days at a time. I think there are some examples where they ran for many weeks. And so I think over time, this is gonna become more and more normal, where the models are running for a very, very long period of time, and you, you don't have to sit there

23. 1:08:38 – 1:11:16

    Pro tips for using Claude Code effectively

    1. BCBoris Cherny1:08:38

       and babysit them anymore.

    2. LRLenny Rachitsky1:08:39

       So we just talked about tips for building AI products. Any tips for someone just using Claude Code, say, for the first time, or just someone already using Claude Code that wants to get better? What are, like, a couple pro tips that you could share?

    3. BCBoris Cherny1:08:51

       I will give a caveat, which is there's no one right way to use Claude Code. So I, I can share some tips, but honestly, this is a dev tool. Developers are all different. Developers have different preferences, they have different environments, so there's just so many ways to use these tools. There's no one right way. Um, you, you sort of have to find your own path. Luckily, you can ask Claude Code. Uh, it's able to make recommendations, it can edit your settings, it kind of knows about itself, so it can help, it can help with that. A few tips that generally I find pretty useful. So number one is just use the most capable model. Um, currently, that's Opus 4.6. I have maximum effort enabled always. The thing that happens is, sometimes people try to use a less expensive model, like Sonnet or something like this, but because it's less intelligent, it actually takes more tokens in the end to do the same task. Um, and so it's actually not obvious that it's cheaper if you use a less expensive model. Often, it's actually cheaper and less token-intensive if you use the most capable model, 'cause it can just do the same thing much faster with less correction, less, uh, less hand-holding, and so on. So that's the first tip, is just use the best model. The second one is use plan mode. I start almost all of my tasks in plan mode, maybe, like, eighty percent. And plan mode is actually really simple. All it is, is we inject one sentence into the model's prompt to say, "Please don't write any code yet." That's it. Like, there's, there's actually, like, nothing fancy going on. It's just the simplest thing.

    4. LRLenny Rachitsky1:10:10

       Mm.

    5. BCBoris Cherny1:10:11

       Um, and so for people that are in the terminal, it's just Shift+Tab twice, and th- that gets you into plan mode. Uh, for people in the desktop app, there's a little button. On web, there's a little button. It's coming pretty soon to mobile also, uh, and we just launched it for the Slack integration, too. Uh, so plan mode is the second one, and, uh, essentially, the model will just go back and forth with you. Once the plan looks good, then you let the model execute. I auto-accept edits after that. 'Cause if the plan looks good, it's just gonna one-shot it. It'll get it right the first time, almost every time with O- Opus 4.6. And then maybe the third tip is just play around with different interfaces. I think a lot of people, when they think about Claude Code, they think about a terminal. Um, and, you know, of course, we support every terminal. We support, like, Mac, Windows, you know, like, whatever terminal you might use, it works perfectly. But we actually support a lot of other form factors, too. Like, you know, we have, like, iOS and Android apps. We have a desktop app. There's, uh, you know, the Slack integration. There's all sorts of things that we support. So I would just, like, play around with these. And again, it's like every engineer is different. Everyone that's building is different. Just find the thing that feels right to you and, and use that. You don't have to use a terminal. It's the same Claude agent running everywhere.

    6. LRLenny Rachitsky1:11:15

       Amazing.

24. 1:11:16 – 1:12:13

    Thoughts on Codex

    1. LRLenny Rachitsky1:11:16

       Okay, just a couple more questions to round things out. What's your take on Codex? How do you feel about that product? How do you feel about where they're going, just kind of competing in this very competitive space, uh, in coding agents?

    2. BCBoris Cherny1:11:30

       Yeah, I actually haven't really used it, but, uh, I, I think I did use it maybe when it came out. It looked a lot like Claude Code to me, so that was kind of flattering. It's... I think it's actually good, you know, to have more competition because people should get to choose, and hopefully it forces all of us to, like, do an even better job. Honestly, for our team, though, we're just focused on solving the problems that users have. Um, so for us, you know, we don't spend a lot of time looking at competing products. We don't really try the other products. I... You know, you kinda, you wanna be aware of them, and you wanna know they exist, but for me, I just, I love talking to users. I love making the product better. Um, I, I love just acting on, on feedback, so it's really just about building a, building a good product.

25. 1:12:13 – 1:14:02

    Boris’s post-AGI plans

    1. LRLenny Rachitsky1:12:13

       Maybe a last question. So I talked to Ben Mann, co-founder of Anthropic, with, with to talk to you about. He had a bunch of suggestions, which I've integrated throughout our chat. One question he had for you is, what's your plan post-AGI? What do you think you're gonna be doing? What's your life like once we hit AGI, whatever that means?

    2. BCBoris Cherny1:12:30

       So before I joined Anthropic, um, I was actually living in rural Japan, and it was, like, a totally different lifestyle. Um, I was, like, the only engineer in the town. I was the only English speaker in the town. It was just, like, a totally different vibe. Like, a, a couple of times a week, I would, like, bike to the farmers market. Uh, and, you know, you, like, bike by, like, rice paddies and stuff. It's just, like, a totally different speed than... Just complete opposite of San Francisco. One of the things that I really liked is, uh, a, a way that we got to know our neighbors, and we kinda built friendships, is by trading, like, pickles.... So in that, in the town where we lived, it was actually like everyone made like miso, everyone made pickles. Uh, and so I actually got like decently good at making miso. Um, and you know, I made a bunch of batches, and, um, this is something that I still make. Uh, miso is this interesting thing where it teaches you to think on these long time scales that's just very different than engineering, 'cause like a, you know, like a batch of white miso, it takes like at least three months to make. And a red miso is like, you know, two, three, four years. You just have to be very patient.

    3. LRLenny Rachitsky1:13:31

       Wow!

    4. BCBoris Cherny1:13:31

       You kind of mix it up, and then you just like wet its head. You have to be very, very patient. So I- the thing that I love about it is just thinking in these long time scales. Uh, and yeah, I think post-AGI, or if I wasn't at Anthropic, I'd probably be making miso. [chuckles]

    5. LRLenny Rachitsky1:13:46

       [chuckles] I love this answer. Uh, Ben asked me to ask you about what's the deal with you and miso? And, uh, so I love that you, [chuckles] answered it.

    6. BCBoris Cherny1:13:54

       [chuckles]

    7. LRLenny Rachitsky1:13:55

       Okay, so the future, the future might be just going deep into miso, getting r- really good at, at making miso. Uh,

26. 1:14:02 – 1:27:44

    Lightning round and final thoughts

    1. LRLenny Rachitsky1:14:02

       amazing. Uh, Boris, this was incredible. I feel like we're, we're brothers now from Ukraine. Uh, before we get to our very exciting lightning round, is there anything else that you wanted to share? Is there anything you want to leave listeners with? Anything you want, uh, you want to double down on?

    2. BCBoris Cherny1:14:18

       Yeah. I, I think I would just like to underscore, you know, like for, for Anthropic, since the beginning, this idea of like starting at coding, then getting to tool use, then getting to computer use has just been the way that we think about things. And we-- this is the way that we know the models are gonna develop or the, you know, the way that we wanna build our models. And it's also the way that we get to learn about safety, study it, and improve it the most. So, you know, everything that's happening right now around, you know, just like c- Claude Code becoming this huge, you know, multi-billion dollar business, and, you know, like now all my friends use Claude Code, and they just text me about it all the time. Uh, so just like, you know, this thing getting kinda big, in, in some ways, it's a total surprise because this isn't kind of the... We didn't know that it would be this product. We didn't know that it would start in a terminal or anything like this. But in some ways, it's just totally unsurprising because this has been our belief as a company for, for a long time. At the same time, it just feels still very early. You know, like most of the world still does not use Claude Code, most of the world still does not use AI. So it, it just feels like this is one percent done, and there is so much more to go.

    3. LRLenny Rachitsky1:15:21

       Oh, man, that's insane to think, seeing the numbers that are coming out. You guys just raised a bazillion dollars. Uh, I think Claude Code alone is making two billion dollars in revenue. You think Anthropic, I think the number you guys put out, you're making fifteen billion in revenue. It's, uh, insane to just think this is- i- how early it still is and just the numbers we're seeing.

    4. BCBoris Cherny1:15:42

       Yeah. Yeah, yeah. It, it's crazy. And, and, and I mean, like the, the way that Claude Code has kept growing is honestly just the users. Like we-- so many people use it, they're so passionate about it. They fall in love with the product, and then they tell us about stuff that doesn't work, stuff that they want. And so, like, the only reason that it keeps improving is because everyone is using it, everyone is talking about it, everyone keeps giving feedback, and this is just the single most important thing. And-

    5. LRLenny Rachitsky1:16:04

       Mm

    6. BCBoris Cherny1:16:04

       ... you know, for me, this is the way that I love to spend my days, just talking to users and making it better for them.

    7. LRLenny Rachitsky1:16:09

       And making miso.

    8. BCBoris Cherny1:16:11

       And making miso. Well, the, you know, the miso is, like, not super involved. It just, you just gotta wait. [chuckles]

    9. LRLenny Rachitsky1:16:16

       You just gotta wait. Well, Boris, with that, we've reached our very exciting lightning round. I've got five questions for you. Are you ready?

    10. BCBoris Cherny1:16:23

        Let's do it.

    11. LRLenny Rachitsky1:16:24

        First question: What are two or three books that you find yourself recommending most to other people?

    12. BCBoris Cherny1:16:29

        I am a big reader. Uh, I would start with a technical book. One is, it, it is Functional Programming in Scala. This is the single best technical book I've ever read. It's very weird because you're probably not gonna use Scala, and I don't know how much this matters in the future now, but there's this just elegance to functional programming and thinking in types, and this is just the way that I code and the way that I can't stop thinking about coding.

    13. LRLenny Rachitsky1:16:51

        Wow!

    14. BCBoris Cherny1:16:51

        So, you know, you could think of it as a historical artifact. You could think of it-

    15. LRLenny Rachitsky1:16:54

        Okay

    16. BCBoris Cherny1:16:54

        ... as something that will level you up.

    17. LRLenny Rachitsky1:16:56

        I love this.

    18. BCBoris Cherny1:16:56

        Uh-

    19. LRLenny Rachitsky1:16:57

        Never before mentioned book, my favorite.

    20. BCBoris Cherny1:16:59

        Oh, amazing. Amazing. Uh, okay, second one is, uh, Accelerando by Stross. This is probably... You know, like, my, my big genre is, uh, is sci-fi, uh, like probably sci-fi and fiction. Accelerando is just this incredible book, and it, it, it's just so fast-paced. The pace gets faster and faster and faster, and I just feel like it captures the essence of this moment that we're in, more than any other book that I've read, just the speed of it. And it starts ch- as liftoff is starting to happen and, you know, starting to approach the singularity, and it ends with, like, this, like, collective lobster consciousness orbiting Jupiter. Um, and it, you know, this happens over, like, the span of a few decades or something. So the, the pace is just incredible. I, I really love it. Maybe I'll, I'll do one more book, uh, The Wandering Earth, uh, Wandering Earth by, uh, Cixin Liu. So he's the guy that did, uh, Three-Body Problem. I think a lot of people know him for that. I actually, I think Three-Body Problem is awesome, but I actually like his short stories even more. So Wandering Earth is one of the short story collections, and he just has some really, really amazing stories. And it, it's also just quite interesting to see, uh, Chinese sci-fi because it has a very different perspective than Western sci-fi, and kind of the way that, um, at least he as a writer thinks about it. So it's just really, really interesting to read and just beautifully written.

    21. LRLenny Rachitsky1:18:15

        It's so interesting how sci-fi has prepared us to think about where things are going, just like it creates these mental models of like: "Okay, I see. I've read about this sort of world."

    22. BCBoris Cherny1:18:24

        Yeah, I think, I think for me, this is, like, the reason that I joined Anthropic, actually, 'cause, uh, you know, like, like I said, I was living in this rural place. I was thinking these long time scales because everything is just so slow out there, at least compared to SF. Um, and just like all the things that you do are based around the seasons, and it's based around this food that takes many, many months. That's the way that kind of, like, social events were organized. That's the way you kind of organize your time. You, like, you go to the farmers market, and it's like, it's persimmon season, and you know that because there's, like, twenty persimmon vendors, and then the next week, the season is done, and it's like grape season, and you kind of see this. So it's like these kind of long time scales. And I was also reading a bunch of sci-fi at the time, and just, like, being in this moment, I was like, you know, just thinking about these long time scales-... I know how this thing can go, and I just, I felt like I had to contribute to it going a little bit better. And that's actually why I ended up at Anth. And Ben Mann was also a big part of that, too.

    23. LRLenny Rachitsky1:19:17

        I feel like I wanna do a whole podcast just talking about your time in Japan, [chuckles] and the journey of Boris-

    24. BCBoris Cherny1:19:22

        [chuckles]

    25. LRLenny Rachitsky1:19:22

        - through Japan to Anthropic. But we'll keep it, we'll keep it short. Uh, I'll quickly recommend a sci-fi book to you if you haven't read it. Have you read Fire Upon the Deep?

    26. BCBoris Cherny1:19:31

        Uh, this is Vinge, right? Yeah, yes.

    27. LRLenny Rachitsky1:19:34

        Yes. Okay. That one's like, it's, like, so interesting from a AI, AGI perspective. Uh, so few people have read that, so, um, I, I love that.

    28. BCBoris Cherny1:19:42

        Yeah, I love it myself.

    29. LRLenny Rachitsky1:19:43

        Yeah. It's like-

    30. BCBoris Cherny1:19:44

        I really like, uh-

Episode duration: 1:27:44