打造 AI 原生工程团队

Hey folks, do y'all hear me okay? 嘿大家，能听到我说话吗？ Okay, I I 好，我 I swear this is not a Claude Code thing, but do you guys mind if I take a photo? 我发誓跟 Claude Code 无关，但能让我拍张照吗？ Cuz cuz Boris and Jared had their session at 2:00 and I really thought this was going to be empty. 因为 Boris 和 Jared 的分享是两点，我真的以为这场会没什么人。 I'm like, there is just no way people would still be coming in from that session. 我心想，那场散场后怎么可能还有人过来。 So, Oh my gosh. 所以，天哪。 [screaming] [尖叫] Thank you, prompt. 谢谢，提示词。 I promise me and Boris don't just do selfie words 我保证我和 Boris 不是只会拍自拍话的人， all the 全部 [laughter] [笑声] But good afternoon and thanks for attending. 不过下午好，感谢大家过来。 So, yeah, my name is Fiona Fung and I lead Claude Code and Cowie engineering and product. 我叫 Fiona Fung，目前负责 Claude Code 和 Cowie 的工程团队和产品。 So, I work really closely with Boris and Cat. 我和 Boris、Cat 合作很紧密。 And before Anthropic, I had led and grown teams at Meta and then also Microsoft. 在 Anthropic 之前，我在 Meta 和 Microsoft 都带过团队、做过增长。 And so, for today's talk, this is kind of like overall agenda, but the whole idea is what are some of the lessons I learned and helping Claude Code and Cowie kind of like grow and as we're building out this team. 今天的分享是整体 agenda，核心是我在帮助 Claude Code 和 Cowie 成长、搭建团队过程中学到的那些教训。 And kind of like what things I and it's it's interesting as lessons I learned even if I think about my time at Meta or even Microsoft, but even Anthropic. 这些教训挺有意思的，回头看在 Meta、Microsoft，甚至 Anthropic 的经历，都有共鸣。 Like it's funny, I did this slide deck maybe like a month ago and also already I've had to change some of the content cuz for example, when I started this deck, there were no routines and that even like that way of working was different for me. 说来也好笑，这套 slides 大概一个月前做的，现在有些内容已经要改了，比如做 slides 的时候还没有 Routines，那种工作方式对我来说也是新的。 And so, yeah, like we really want to like I want to kind of cover five themes that I've noticed. 所以，我想聊五个我注意到的主题。 One which is the bottlenecks have moved, they've shifted. 第一，瓶颈移位了。 And so, when bottlenecks shift, what are then some of the team norms that we had to rewrite within the Claude Code team? 瓶颈移位之后，Claude Code 团队有哪些团队规范是我们不得不重写的？ I also wanted to share a little bit about all these team norms we had to rewrite, how we rolled them out, and also what are some of the proof as some of the you know, like some some signals that I get of well, yeah, we're trending in the right direction. 我还想聊聊这些规范是怎么推出去的，以及我拿到的哪些信号让我觉得，对，我们方向对了。 And it's always going to be important for me to kind of look at to see is it still serving us, trending in the right direction? 这对我来说始终很重要：持续看，它还在帮我们，还在朝正确方向走吗？ And then I'll end it with a few kind of like questions that I still have for myself, and then also some suggestions for you to maybe take an action and embark uh to your teams to have conversations together. 最后我还会分享一些我自己还没想清楚的问题，以及一些建议，大家可以拿回去跟自己的团队一起聊。 So with that, the first section, the bottlenecks have moved. 好，第一节，瓶颈移位了。 I call it the shift. 我叫它「转变」。 But you'll probably hear me repeat this kind of subtitle text a lot, which is what served you prior may not serve you any longer. 你们大概会听我反复提一句 slide 上的副标题：以前管用的，现在未必管用了。 And when actually even when I think about all my experience, whether it was like at Anthropic or Meta or Microsoft, the constant growth mindset is just a muscle that has served me really well. 回顾我在 Anthropic、Meta、Microsoft 的所有经历，持续的成长型思维是一块肌肉，一直很好用。 And especially right now when the rate of I don't know if y'all are feeling it, the rate of change is just a little bit crazy, right? 尤其是现在，不知道大家有没有感受到，变化的速度真的有点疯狂，对吧？ Like I remember the first time I started doing some live coding was last year, and it was still making some some, you know, bugs that I'm like, "Ah, why why are you using constants everywhere? 我记得第一次做现场 coding 是去年，当时还是会出一些 bug，我就想：哎，为什么到处用常量啊？ That's not good engineering practice." 这可不是好的工程实践。 And now it's it's, you know, just become so much more capable. 而现在它已经变得强大太多了。 But that's that's what I I'm seeing as this interesting shift of when the bottlenecks moved, how do you think about adapting uh in terms of everything else around that bottleneck? 但这就是我觉得有意思的转变，瓶颈移位之后，你要怎么调整围绕它的一切？ So maybe y'all are feeling this too, but like for years engineering bandwidth was the expensive thing. 也许大家也有这种感受，很多年来，工程产能都是最贵的东西。 Like coding throughput was really expensive. 代码吞吐量真的很贵。 And when you think about all the processes we have of shipping software, a lot of it was around 当你想到我们交付软件的所有流程，很多都是围绕着 Hi, welcome. 嗨，欢迎！ There's a chair. 那边有椅子。 Take away those reserved tags. 把那些预留位空出来。 There's no one sitting. 没人坐着呢。 [laughter] [笑声] And there's another one right here, sir. 这边也有一个，先生。 Welcome. 欢迎。 Um but yeah, like all of that, like even when you think about how we used to do planning, like remember we used to do like waterfall and then agile, everything was used to be because engineering bandwidth was really expensive. 对，就说这个，比如我们以前怎么做规划，记得吗，先是瀑布流，然后是敏捷，一切都是因为工程产能真的太贵了。 Actually, I'll take a little segue here. 这里我想插一段。 This is not the first time our Like when you think about our industry, we've always had to adapt. 我们这个行业不是第一次面对这种局面了，一直以来都得适应。 Like I'm going to put you all in a time machine. 让我带大家坐时光机回去。 Come back with me all the way to the year 2000s. 跟我一起回到2000年代。 That was when I started my career. 那是我入行的时候。 I worked on Visual Studio, and we were shipping Visual Studio 2005. 我那时候在 Visual Studio 团队，正在交付 Visual Studio 2005 的版本。 And I kid you not, in those days, if folks remember, we used to ship software on like CD-ROMs. 不骗你们，那个年代，大家可能还记得，我们发软件靠的是 CD-ROM。 Before CD-ROMs, it was actually floppy disk. CD-ROM 之前，用的还是软盘。 And so, I still remember VS 2005, there were really hard deadlines. 我至今还记得 Visual Studio 2005，那时 deadline 真的死死的。 We had to hit those deadlines cuz we had to get the software to the manufacturing lab to print on the CDs, to put in the boxes, to ship in the stores. 必须赶上，因为软件要送去制造厂压盘，装进盒子，再发到门店。 And so, when you when you even think about that, when we were able to distribute software online, that also changed how we ship software. 后来我们能在线分发软件了，交付方式也跟着变了。 And so, that's what I'm finding really interesting of this this new, you know, shift that I'm seeing is engineering bandwidth. 所以我觉得最有意思的，就是我正在经历的这个新转变，工程产能这件事。 It's no longer the expensive thing. 它不再是最贵的东西了。 So, for example, on the Cloud Code team, for sure, coding is rarely the slow part anymore. 就拿 Claude Code 团队来说，代码本身几乎已经不是慢点了。 Um and I would say it's not even that it's not the slow part, it's just also the throughput has really, really increased. 而且我想说，不只是它不慢了，吞吐量也真的大幅提升了。 So, it's not only like, "Yay, we're all all getting to build more." 所以不只是「大家都能做更多了」这么简单。 It's just the amount that we're generating has also changed a lot. 我们产出的代码量本身，也变了很多。 And so, what we saw was, you know, when your your bottleneck shifts from kind of the coding and the actual act of typing, like if you remember, it used to be writing code was expensive or writing tests or refactoring. 你知道，当瓶颈从写代码这件事本身移开，像是记得吗，以前写代码很贵，写测试、做重构都很贵。 I remember all of these conversations of, "We have to schedule some time to do refactoring. 我还记得那时的对话：“我们得专门排时间做重构。 Oh, but we have to do product work, and this is expensive. 哦，但产品需求也要做，时间根本不够用。 When are you going to find that time to do it?" 那时间从哪儿来？” All of that has shifted now. 这一切现在都变了。 That is no longer the bottleneck. 那已经不再是瓶颈了。 And so, when that happened, sometimes I notice the bottlenecks end up shifting towards other areas. 所以，当这个变化发生后，我发现瓶颈往往会转移到其他地方。 And so, what happened? 那么，后来发生了什么？ Like, for example, verification, review, cross-functional partners, security. 比方说，验证、review、跨职能伙伴合作、安全。 Because coding is no longer the bottleneck, and also we're doing so much more of it, these are some of the new bottlenecks that we're seeing. 写代码不再是瓶颈，而且我们写的代码量大了很多，这些就成了我们正在面对的新瓶颈。 So, it's really always us asking, "Is this code correct? 所以我们一直在问的问题是：“这代码对吗？ Who reviews this code?" 谁来 review 这些代码？” That's like probably one of the top questions I get from all fellow end leaders. 这大概是我从所有工程 leader 那里收到最多的问题之一。 Like, "How are humans keeping up with how you guys are doing like code reviews. “人类怎么跟上你们做 code review 的节奏？” And interestingly, also how is it maintained? 还有一个有意思的问题：代码怎么维护？ Because now it is also a lot easier for us all to generate a lot of code. 因为现在大家要生成大量代码，变得容易多了。 So also thinking about that maintenance cost, too. 所以也要考虑维护成本这件事。 So with that, these were some of the processes that I noticed quietly stops working. 正是这些，是我发现悄悄失效的那些流程。 And I love that phrase quietly stops working cuz I don't know if you all like a lot of times we all put in processes hopefully for a reason, right? 我特别喜欢“悄悄失效”这个说法，因为你知道，很多时候我们建立流程都是有原因的，对吧？ Like we're thinking, "Hey, there was a gap here or we want to improve." “嗯，这里有个空缺，或者我们想改进某件事。” But what I've found over the years is rarely do processes kill themselves. 但我这些年观察下来，流程几乎从不会自己消亡。 We tend to just layer more and more and more processes on. 我们往往是一层又一层地往上叠新流程。 Like I remember that on one team, we had so many SLAs. 我记得有个团队，我们有好多 SLA。 Like there was a P0 bug SLA, a high pri- like sub review instead of like 比如 P0 bug SLA、高优先级子任务 review SLA，诸如此类， Anyways, there's so many SLAs. 反正 SLA 多得数不清。 After a while I'm like, "Oh no, we need to stack rank priorities so that all the engineers knew which SLA was going to be even more important." 到后来我说：“不行，我们得给优先级排个序，让大家知道哪条 SLA 才是最重要的。” And I remember even at that time I thought, "Hey, we should start thinking about what things we should defrag a little bit." 我记得那时我就想：“我们应该开始想想，哪些东西可以稍微精简一下。” But um yeah, so these are the processes that might have served you. 嗯，这些就是那些可能曾经管用的流程。 Do you remember again is that line of what may have served you may not serve you any longer. 记住那句话：以前管用的，现在未必管用了。 But like the planning norms. 先说规划规范。 We used to spend a lot more time, you know, pre-planning because coding time was expensive. 我们以前要花更多时间提前做规划，因为写代码的成本很贵。 Uh code ownership, there used to also be a lot of questions of who who who wrote this code? 代码归属，以前也有好多问题：这段代码到底是谁写的？ Who owns it? 归谁负责？ That's a little bit of an other question now. 现在这个问题已经不那么重要了。 Uh code reviews what we'll get into a little bit. 代码 review，我们等会儿会讲到。 Team makeup is interesting, too. 团队构成也很有意思。 It's not only like roles are blurring, right? 不只是角色在模糊，对吧？ Like so between like engineers can also 工程师们也可以 So now now have AI to help augment non-engineering roles. 现在有 AI 来帮非工程角色增效了。