Anthropic 的 Boris Cherny：为什么说编程已被解决，下一步是什么

Okay, I'm excited to introduce our next speaker. 好，我很高兴介绍我们的下一位演讲嘉宾。 Show of hands, who here uses Claude code? 请举手，在座有谁在用 Claude Code？ Okay, show of hands, who here has Claude code psychosis? 好，请举手，在座有谁有 Claude Code 沉迷症？ Come on guys, [clears throat] it's okay. 来吧大家，没事的。 It's okay. 没事的。 Um my my my team lovingly says I have Claude code psychosis, which may or may not be true. 我的团队亲切地说我有 Claude Code 沉迷症，这可能是真的，也可能不是。 Um we are delighted to have Boris Cherny with us today. 我们非常高兴今天有 Boris Cherny 来到现场。 Uh Boris is the creator, the father of Claude code. Boris 是 Claude Code 的创造者，它的父亲。 Um and uh in the process of doing that has just had a front row seat to to reinventing uh the modern way of of software development. 在这个过程中，他亲眼见证了现代软件开发方式的重新发明。 Um and we're we're really grateful to you, Boris, for taking the time to speak with us today. 我们非常感谢你，Boris，今天抽出时间来和我们交流。 We know that um the entirety of software development kind of rests on your shoulders. 我们知道整个软件开发的重担可以说都压在你肩上。 So, thank you for taking it out of your time to be with us today. 感谢你抽出时间来陪我们。 And interviewing Boris is Lauren Reader from our team. 采访 Boris 的是我们团队的 Lauren Reeder。 Thank you. 谢谢。 Giving our chairs. 换座位了。 Um you took my you took my opening line, Asia. 你抢了我的开场白，Sonya。 We asked who here uses Claude code. 我们问过在座谁在用 Claude Code。 There's a lot of hands. 举手的人还挺多。 That's awesome. 太棒了。 Thank you for joining us, Boris. 谢谢你来，Boris。 It's very special to have you here. 很高兴你能来。 Um as a roomful of builders, I think you are changing building entirely. 在座都是开发者，我觉得你正在彻底改变构建这件事本身。 And so, I'm very curious to explore how you think about the future of software, coding, and what we should spend all of our free time on. 所以我很想聊聊你对软件未来、编程的看法，以及我们应该把多出来的时间花在哪里。 Um but I'll give you a me a tiny bit more background on you so that everyone has a little bit more context. 我先给大家多介绍一点你的背景，让大家有更多了解。 So, beyond creating Claude code, Boris is very much an engineer's engineer. 除了创建 Claude Code，Boris 是一位非常纯粹的工程师的工程师。 You were writing a lot of code through your whole career, writing textbooks about code, including programming in TypeScript. 你整个职业生涯都在大量写代码，还写过关于代码的教材，包括 TypeScript 编程。 Um and I think last time we chatted you hadn't written a single line of code in the last year, or at least so far in 2026, which is quite the change. 我记得我们上次聊天时，你在过去一年里，或者至少 2026 年到现在，一行代码都没写过，这是个很大的转变。 Um 嗯 There's also a a little known thing back in middle school, I wrote a guide about uh writing BASIC for TI-83 Plus calculators. 还有一件鲜为人知的事，初中的时候，我写过一个 TI-83 Plus 计算器 BASIC 编程指南。 And I I just I I searched for it, it's actually still on the internet. 我刚搜了一下，现在网上居然还能找到。 It's extremely embarrassing, so please don't search it. 太尴尬了，千万别去搜。 But it [laughter] exists. 但它确实存在。 We will definitely be finding that. 我们肯定会去找的。 Um so, we're going to do I'm going to start with a few questions here. 好，我先来问几个问题。 Maybe we'll start with a little bit of the history of Claude code, how you started it, and then we're going to have a lot of audience Q&A for this one. 先聊一聊 Claude Code 的历史，你怎么开始做这个的，然后我们会留很多时间给大家提问。 And so, start thinking about your questions in the back of your head, uh and would love to turn it over to you all soon. 大家现在可以开始想问题了，很快就会把时间交给你们。 Yeah. 好。 Um and also real quick, so for people that use Claude code, do people use the CLI mostly? 对了快问一下，用 Claude Code 的人，大家主要用 CLI 吗？ Like okay, majority CLI? 好，大多数用 CLI？ Okay. 好的。 That's a lot. 还挺多的。 Majority desktop? 大多数用桌面端？ Okay. 好。 Majority VS code or JetBrains IDE? 大多数用 VS Code 还是 JetBrains IDE？ Okay. 好。 That's actually not a lot. 那其实不多。 Okay. 好。 Other? 其他？ I'm like iOS mostly these days. 我现在主要用 iOS。 Yeah. 是的。 [laughter] [laughter] Okay. 好。 Cool. 好的。 Um yeah, so I started Claude code kind of accidentally in a in a lot of ways. 我做 Claude Code 某种程度上是误打误撞。 Um I joined this team back in late 2024. 我是 2024 年底加入这个团队的。 It was a sort of this incubator within Anthropic called Anthropic Labs. 那是 Anthropic 内部的一个孵化器，叫 Anthropic Labs。 And uh the team kind of served its purpose. 这个团队完成了它的使命。 Um we created Claude code, uh MCP, and the desktop app. 我们做出了 Claude Code、MCP 还有桌面应用。 It was a team it was just a few of us. 团队规模很小，就我们几个人。 So, very much like innovation team. 非常像创新小分队。 We built the thing that we wanted to build, we disbanded the team. 我们把想做的东西做出来了，然后团队就解散了。 Uh now the team's actually back together for round two. 现在团队其实又聚回来了，开始第二轮。 Mike Krieger, who's the you know, like the chief product officer at at Anthropic and used to be one of the founders at Instagram, so he's leading that right now. Mike Krieger，他是 Anthropic 的首席产品官，之前是 Instagram 的创始人之一，现在他在带这个。 Um so the kind of the the the the reason that I started to work on coding is we felt like there was this product overhang. 我当时开始做编程这个方向，是因为我们觉得存在产品层面的欠账，模型能力已经超前，但产品还没跟上。 And I I'm guessing people here use that word a lot. 我猜在场的人应该经常听到这个词。 Uh but we definitely use this word a lot in kind of within the lab. 但我们在实验室内部确实很常用这个词。 Uh there's this idea that the model can do all the stuff that no product has yet captured. 意思是模型能做到的事情，还没有任何产品真正把它发挥出来。 And in late 2024, when we were looking at coding, the way that we did coding, the state of the art at the time was type ahead. 2024 年底我们研究编程方向时，当时最先进的做法是 type ahead。 It was you open your IDE and you press tab and you can like complete like one line at a time. 就是你打开 IDE，按一下 tab，一次补全一行代码。 And that was the thing that Sonnet 3.5 enabled for the first time. 而 Sonnet 3.5 是第一次让这件事成为可能。 But the feeling was we could actually go a lot further than that. 但我们的感觉是，可以走得比这远得多。 And the model was almost ready for the next big step. 模型已经差不多准备好迈出下一步了。 So, we don't have to do type ahead anymore, we can just have the agent write all of the code. 所以我们不需要再做 type ahead 了，可以直接让 agent 把所有代码都写出来。 And so, I built it, and it just really didn't work for the first 6 months. 然后我就去做了，但头 6 个月真的完全行不通。 It was like not very good. 效果不咋样。 It was barely usable. 基本上没法用。 I wrote it from I used it for maybe 10% of my code or something like that. 我自己大概只用它来写 10% 的代码。 And even after we released Claude code initially, it was not a hit. 就算 Claude Code 最初发布之后，也没有火起来。 There's a lot of people that used it, but it did not have this exponential growth that it has today. 用的人不少，但没有像今天这种指数级增长。 Um that started with Opus 4 in May. 那是从 5 月份的 Opus 4 开始的。 And I I remember that very clearly. 我对那个时刻记得很清楚。 That's like when the exponential growth started, and then it kind of inflected with every model release. 就是从那时候开始指数增长，然后每次模型发布都会再往上拐一下。 Uh like it started with Opus 4, then 4.5, then 4.6, now 4.7. 从 Opus 4 开始，然后是 4.5，4.6，现在是 4.7。 It just kind of keeps inflecting. 就这样一直在往上拐。 But essentially, we were trying to build this thing that was like pre-PMF, and we knew that it wouldn't have PMF for 6 months because we were building for the next model. 本质上，我们当时在做的东西还是 pre-PMF 阶段，我们也知道它在 6 个月内不会有 PMF，因为我们是在为下一代模型做准备。 And that was the idea the pretty much the whole time. 这基本上从头到尾都是核心思路。 And you know, for Anthropic in general, we've always just been very focused. 对 Anthropic 来说，我们一直都很专注。 We've always cared about business and enterprise and safety and coding. 我们一直都很在意商业、企业、安全和编程这几个方向。 That's just always been kind of the way that we wanted to build. 这就是我们一直想走的路。 And so, at some point we kind of knew that we wanted to build a product. 所以到某个时间点，我们大概就知道自己想做一个产品了。 We didn't know exactly what we wanted. 我们当时也不知道具体想做什么。 So, this kind of ended up being the the product bet. 最后就押在了这个产品方向上。 It's an incredible story, especially that it was an accident. 这故事太精彩了，尤其是它居然是个意外。 Um so, you've said on the record that you think coding is solved. 你公开说过，你觉得编程这件事已经被解决了。 Uh if this is one of the three best from Anthropic, can you tell us more about what you mean by that, and what might still not be solved, or what second-order problems might come? 如果这是 Anthropic 最出色的三件事之一，能跟我们说说你的意思吗？还有哪些还没解决，或者会带来哪些二阶问题？ All right. 好。 I can ask another question for the room. 我可以再问现场的人一个问题。 Who writes 100% of their code by hand? 谁是 100% 手写代码的？ Who writes 100% of their code using a agent like Claude code? 谁是 100% 用 Claude Code 这类 agent 写代码的？ Okay. 好。