GitHub 的 Agent 时代：提交量 14 倍增长、2 亿开发者、Copilot 下一步 — Kyle Daigle

I just find that the folks that came from a different career, went to school for something else, went off and did this random thing and then became a software dev or were a dev, did a random thing, came back, learning that extra set of information, learning those extra skills and now having the power of an AI where I can crank up 15 agents on Saturday, you know, while my kids are doing lacrosse. 我就是觉得，那些从别的行业转过来、学的是其他专业、做了些乱七八糟的事才成为开发者的人，或者本来就是开发者、去做了别的事再回来的人，带着那份额外的积累、那些额外的技能，现在又有了 AI 的加持，我周六可以同时跑 15 个 Agent，你知道，就在孩子们打曲棍球的时候。 Uh that's like really powerful and I think it gets me back to that feeling of like creation and it's very hard to like replicate that in most other senses. 那种感觉真的很有力量，我觉得它让我找回了创造的感觉，而那种感觉在其他大多数事情上都很难复制。 Before we get into today's episode, I just have a small message for listeners. 在进入今天的节目之前，我想给听众们说几句话。 Thank you. 谢谢你们。 We would not be able to bring you the AI engineering, science, and entertainment content that you so clearly want if you didn't choose to also click in and tune into our content. 如果不是你们选择点进来、收听我们的内容，我们根本无法给你们带来这些你们明显想要的 AI 工程、科学和娱乐内容。 We've been approached by sponsors on an almost daily basis. 几乎每天都有赞助商来找我们。 But fortunately, enough of you actually subscribe to us to keep all this sustainable without ads and we want to keep it that way. 不过幸运的是，订阅我们的人足够多，让我们不靠广告也能维持下去，我们希望一直保持这样。 But I just have one favor to ask all of you. 但我只想请大家帮一个忙。 The single most powerful, completely free thing you can do is to click that subscribe button. 你能做的最有力、完全免费的一件事，就是点击订阅按钮。 It's the only thing I'll ever ask of you. 这是我唯一会向你们开口的事。 And it means absolutely everything to me and my team that works so hard to bring the Inspace to you each and every week. 这对我和我的团队意义重大，他们每周都在拼命工作，把 Latent Space 带给你们。 If you do it, I promise you we'll never stop working to make the show even better. 如果你们订阅了，我保证我们永远不会停止让节目变得更好。 Now, let's get into it. 好，开始吧。 Okay, we're here with Cargo, CEO of GitHub. 好的，今天我们请到了 GitHub 的 COO Kyle Daigle。 Welcome. 欢迎。 Yeah, thanks for having me. 谢谢邀请。 You're not just CEO of GitHub. 你不只是 GitHub 的 CEO。 People know you as that. 大家都这么认识你。 Yeah, 对， you have a new role. 你还有个新职位。 Yeah, so I have an expanded role now. 对，我现在职责扩大了。 I mean we've been working uh I've been working at GitHub for 13 years and doing uh you know all things developer joined as a developer myself and now uh I'm also responsible as the CMO of developer for Microsoft and so all the kind of learnings and passion for developers and how we work with them and how we communicate and uh you know how we bring our products to market 我是说，我在 GitHub 工作了 13 年，一直做开发者相关的事情，以开发者身份入职，现在我还兼任 Microsoft 的开发者 CMO，把这些年对开发者的了解、热情，以及怎么跟他们合作、怎么沟通、怎么把产品推向市场， we're also bringing that expertise you know to the broader Microsoft ecosystem and uh and helping uh every developer that uses a Microsoft, you know, product or would like to uh to have a sort of similar experience that they've had with, you know, GitHub uh over the years. 这些经验也带到了更广泛的 Microsoft 生态系统，帮助每一个使用 Microsoft 产品的开发者，或者希望体验 GitHub 这些年所提供的那种感觉的开发者。 So, it's a big different uh role in some ways, but it's also just building on the experience that, you know, I've had at GitHub of just sort of tell the truth, be authentic, show people how to use it, and then let the you know, products speak for themselves. 所以从某种程度上说是个很不同的新角色，但也只是在 GitHub 的经验上继续延伸，就是说实话、真实，告诉大家怎么用，然后让产品自己说话。 Not just doing that with all of Microsoft. 现在只是把这些也延伸到整个 Microsoft。 Yeah. 对。 And uh we're we'll be releasing this in conjunction with Build. 我们会配合 Build 大会同步发布这期内容。 you got lots of stuff planned and we can sort of touch on that whenever it's appropriate. 你们有很多东西要公布，我们可以在合适的时候聊聊。 Uh I think one of the interesting things is I rarely meet a COO who's also a CMO. 我觉得有意思的是，我很少见到同时担任 COO 又是 CMO 的人。 I think you're a very 我觉得你是个非常 outward facing and you're very confident publicly. 面向外部、公开场合非常自信。 Um that's rare. 嗯，这很少见。 Like do you actually view yourself as COO? 你真的把自己定位成 COO 吗？ Like what's 你是怎么看 Yeah, I mean 是啊，我是说 what is your thing? 你最认同哪个身份？ I think for me like it's been funny. 我觉得对我来说这挺有意思的。 The titles have always been uh like always felt a little strange to me. 这些头衔对我来说一直有点怪。 I mean, I joined GitHub as a developer, you know. 我是说，我当年是以开发者身份加入 GitHub 的。 I mean, I wrote so much of the 我是说，我写了很多 Let's bring that up. 说说这个。 Yeah. 对。 Yeah. 对。 You wrote the back end. 你写了后端。 Yeah. 对。 I was going through like uh I was going through uh some old photos uh when, you know, folks were talking about, you know, how things were being built and how others would build GitHub. 我当时在翻一些老照片，那时候大家都在聊 GitHub 是怎么搭建的、别人会怎么搭建 GitHub。 I built uh uh web hooks and worked with teams building the API, built the platform layer, anything that integrated with GitHub. 我搭建了 webhooks，参与了 API 团队的工作，做了平台层，凡是跟 GitHub 集成的东西基本上都碰过。 uh up until really 2018 I uh was built or ran the engineering teams uh and that's kind of where my like the beginning of my passion always was was helping people build things deliver them to like their customers. 大概到 2018 年之前，我一直在做工程，领着工程团队，那是我热情最初始的地方，就是帮大家把东西做出来、交付给他们的用户。 Uh and so being a developer building for developers was always super unique and I think as my role expanded it became you know my ability to talk to not just developers but also enterprise customers or you know business leaders and have this like translation layer. 作为开发者，为开发者做产品，这件事一直很独特。后来随着职责扩展，我能跟开发者聊，也能跟企业客户、业务决策层聊，中间有了这个翻译层。 Uh and then through all those years, GitHub has always operated pretty uniquely. 然后这些年来，GitHub 一直以一种很独特的方式运作。 Like post- pandemic, working remotely was not as novel as it was when, you know, GitHub uh uh you know, started in 2008. 比如疫情之后远程工作不再新鲜了，但对 GitHub 来说，2008 年创立时远程就是常态。 But all that expertise of running remote teams, doing it well, became this sort of bigger role ultimately turning into the COO role of how do we operate GitHub in the way that GitHub's always operated after the Microsoft acquisition. 但那套管理远程团队、做好远程的经验，最终演变成了更大的职责，也就是 Microsoft 收购之后，怎么按照 GitHub 一直以来的方式运营 GitHub，这成了 COO 的角色。 Yeah. 对。 uh and kind of so on from there. 之后就顺势延伸下去了。 So I mean like for me I think the I've I still code. 所以对我来说，我觉得，我还是在写代码。 I love coding but the problem has always been like people. 我热爱写代码，但真正难的问题一直是人。 It's a much harder problem to both support our own employees harder problem to communicate to developers and enterprise buyers what we're building why it matters because those are two very different messages. 既要支持自己的员工，又要向开发者和企业买家说清楚我们在做什么、为什么重要，这两套话语体系完全不同，这才是更难的问题。 And so getting to work in the mix of COO, CMO, also just being a dev, uh I think is what's kept me at GitHub for for so long. 所以能在 COO 和 CMO 之间穿梭，同时还能写代码，我觉得这就是我在 GitHub 一待这么久的原因。 Yeah. 对。 Apparently you have your commits have gone up. 你的提交记录好像多了很多。 Yeah. 对。 Uh what's this? 这是怎么回事？ What's going on? 发生了什么？ Yeah. 是啊。 I mean called me out pretty uh pretty aggressively. 我被当众点名，有点激烈。 So I mean you know I think I mean as you can imagine, right? 所以，你能想象，对吧？ Like you can see my like normal era of being a dev in the you know 2013 2014 era and then moving into management and then ultimately the COO role. 你能看到我 2013、2014 年那段真正当开发者的时期，然后进入管理层，最后到 COO。 I think what you see there is me like really getting back to coding thanks to AI. 我觉得你看到的是，多亏了 AI，我真的重新捡起了写代码这件事。 You know, uh I similar to like attaching problems between, you know, how to market and how to operate a business and how to code. 就像我把营销、运营公司和写代码这些问题连接起来一样。 I find like building agents and workflows that are connecting very disperate problems to be what's driving this. 我发现构建 Agent、搭建工作流来连接各种差异很大的问题，正是驱动我的东西。 So it's like some of it's writing software, a lot of it is like connecting a ton of different data sources to like help me out. 有一部分是写代码，但很多是把大量不同的数据源连起来，帮我更好地工作。 Uh uh but that is completely me you know uh really really diving in uh on the AI side um in trying out our tools trying out everyone's tools like um but building for me building for the like non-technical leader though I'm technical you know and how we're uh able to use these tools more than just the simple like call and response that I think you know a lot of the like non-technical your employers like you have to get you have to use AI and so everyone uses like chatpt or copilot or claude or whatever to really get into like how is this going to help me out it 但那完全是我自己深度扎进 AI 这一侧，试用我们的工具，也试用所有人的工具，但都是在为我自己，或者说为那种虽然技术背景够用但不算纯技术的领导者构建，就是说，怎么不只是简单地问答调用，真正用起来帮助自己。 I find that it's not the I need to write a blog post I need to you know those simple examples helping people find the workflows of like okay I need you to go through all the PRs today I need you to go through everything that we've posted online I need you to go through what we've did the last you know three months go through all of my Obsidian notes for any mentions of this then go through my transcripts at work. 我发现最有价值的不是"帮我写篇博客"这种简单用法，而是帮人找到这种工作流：好，帮我过一遍今天所有的 PR，帮我过一遍我们线上发布的所有内容，帮我回顾过去三个月，翻我的 Obsidian 笔记找所有相关提及，再翻工作里的转录记录。 We uh use uh teams. 我们用的是 Teams。 So like using work IQ, go call that MCP server, grab all the transcripts, go through all of Slack, and then build me out the plan of like what this week's messaging actually was. 用 WorkIQ 调用那个 MCP server，拉取所有转录，过一遍所有 Slack 记录，然后整理出本周的沟通重点到底是什么。 That's something that was like impossible because for me, I find AI in like what most of this like launch here is is actually like less building forward. 这件事之前几乎不可能完成，因为对我来说，AI 在大多数这类场景里，其实不是往前走， It's actually like a recursive loop backwards. 而是往回追溯的循环。 I'm always looking at what had happened first. 我一直是先回头看发生了什么。 Like go back through the week and tell me what we did, what worked, what didn't work, you know, 回顾这一周，告诉我做了什么、什么有效、什么没效， and then tell me in the next, you know, three or four days, 然后告诉我接下来三四天， what would you tweak based on, you know, this sort of like looking backwards and then looking ahead a little bit. 根据这种向后看再向前望的方式，你会调整哪些东西。 I find that to be so much more valuable especially for like non-technical because that retrospection is actually very LM are very good at that you know like finding all the patterns pulling them out and then applying that retrospection to just a couple of days or just like a short period of time 我觉得这对非技术型人来说价值极高，因为这种回溯正是语言模型擅长的，找规律、提取出来，然后把回溯应用到接下来几天或者一小段时间里。 uh is all a bunch of apps that I've built and launched like a bunch of uh internal tools 还有一堆我自己搭建并上线的小应用，很多内部工具。 I use the new uh GitHub copilot app the desktop app with workflows every time I crack open my laptop it's running workflows for me 我用新的 GitHub Copilot 桌面应用，每次打开电脑都跑着工作流。 Uh it's just a ton of different stuff and of course it all ends up on it all ends up on GitHub. 还有各种各样的东西，当然最后都托管在 GitHub 上。 Uh of course that's where that's where uh stuff is hosted. 当然，这才是放东西的地方。 Uh man there's so much to ask you. 真的，有太多想问你的。 Uh I was going to leave the the how do you run a company with AI thing at the end. 我本来想把"怎么用 AI 管公司"这个话题留到最后。 I have to ask one one double click one thing. 我要先问一个问题深挖一下。 You said like you're looking back at the week you're you're understanding what uh what happens when you say we 你说你在回顾这一周，理解发生了什么，你说的"我们" that's 3,000 people. 是 3000 人。 Yeah. 对。 Yeah. 对。 how 怎么 I mean I think you know uh when we started rolling out AI internally beyond engineering right um one of the things that I was really really uh passionate about is like we have to do this in a way where no one has to change how they work 我是说，当我们在工程团队之外开始内部推广 AI 的时候，我当时非常在意的一点是，我们要以这种方式来推，不能让任何人改变他们的工作方式。 I don't want to have to teach you a tool 我不想教你用某个工具。 I don't want to have to teach you something new and so for us we tried out a few tools most of them don't work because I got to get you on board 我不想让你去学什么新东西，所以我们试了几款工具，大多数都不行，因为要让你接受就得先让你上手。 you know I got to teach you how to use it what we've actually ended up doing is we've built like a set of 你知道，我要教你怎么用，结果我们实际上做的是搭了一套 uh uh you know skills internally. 内部的 skills。 We have like we each have our set of skills and we've just been distributing even to the nontechnical folks the CLI and then effectively we're just giving it access to like read about everything that we're writing. 我们每个人都有自己那套 skills，然后我们就把 CLI 发给非技术同事，实际上就是让它能读取我们所有写下来的内容。