Why Claude can’t be your PM (yet) | Anthropic CPO Panel

频道: Lenny's Podcast
视频: https://www.youtube.com/watch?v=sEXdyK6woKU
原文语言: en
统计: 共 54 轮 · Dan 16 · Ami 15 · Mike 10

[0:06] Dan

Welcome. How's it going? Good.

欢迎各位。大家怎么样？挺好。

[0:07] Ami

Doing great. How are you all?

好得很。你们大家怎么样？

[0:09] Ami

So nice to see you all. [cheering]

见到大家真开心。（欢呼）

[0:11] Ami

Isn't this a great event? I feel like it's just it's so happy.

这活动是不是特别棒？我感觉就是——气氛特别欢乐。

[0:15]

Yeah. It's good vibes.

是啊，氛围很好。

[0:16]

Yeah. So, I want to start this conversation off with the question that I think is on everybody's minds, which is a year ago we uh were told that AI might make PMs obsolete [laughter] and now I feel like the the vibes are that everyone needs to become a PM. You know, one of the things I love about getting to talk to both of you is just by being an anthropic and working with the models in the way that you do, you have a little bit of like a sense of where the future is going, especially for different job roles. And I'd really love for for both of you to talk to us about what is a great PM at Enthropic look like. Is the PM role going away? Is it changing? And if so, how? I think the job of product is always to be a bridge between the real problems that people have in the world and the technology that you can use to solve it. And like that's always been true. And I think a thing that is weird right now is that it used to be that the technology would change like every 10 years or 5 years and now it feels like it changes like every two months. And that's really hard. That's just like a like as a human I feel like it's really hard to learn how to deal with that sort of change. And so what I always think about is like the technology side is changing but the human side like the human side doesn't change that fast. We have the same problems. And so what I always think about is like how can I be obsessed with like solving the human problems and then stay super fresheyed about how the technology works? like how can I basically throw away most of what I know about what used to work to solve this problem and then try again like knowing everything I do about the problem.

Dan： 对。那我想用一个我觉得所有人心里都在琢磨的问题来开场：一年前，有人跟我们说 AI 可能会让 PM（产品经理）过时（笑），而现在我感觉风向变成了人人都得变成 PM。能跟你们两位聊，我特别喜欢的一点是，你们身在 Anthropic，又是那样跟模型打交道的，所以对未来往哪儿走多少有点感觉，尤其是对各种岗位角色。我特别想请两位跟我们讲讲：在 Anthropic，一个出色的 PM 是什么样的？PM 这个角色是在消失吗？是在变化吗？如果在变，是怎么变的？

Ami： 我觉得产品这份工作，一直都是在人们在现实世界里遇到的真实问题，和你能拿来解决它的技术之间搭一座桥。这一点从来没变过。而眼下有点奇怪的地方在于，过去技术大概每 10 年、每 5 年变一次，现在感觉是每两个月就变一次。这真的很难。就是说，作为一个人，我觉得要学会应对这种变化真的很难。所以我一直在想的是：技术这一侧在变，但人这一侧——人这一侧变得没那么快。我们面对的还是同样的问题。所以我一直在想的是，我怎么才能一门心思扑在解决人的问题上，同时对技术怎么运作始终保持一双新鲜的眼睛？也就是说，我怎么才能把我所知道的那些「过去解决这个问题管用的办法」基本都扔掉，然后带着我对这个问题本身的全部了解，重新再来一遍。

[1:59] Ami

And so I feel like that part hasn't changed. Um, what I think has changed is it's almost like rewinding the clock. Like when I back when the dinosaurs were in the earth and I started in product, like we didn't know, we didn't have definitions. Like we just didn't we didn't know what the job was. And so you were just like a general problem solver and you'd like walk around and try to solve a problem until you like ran into a wall and then you would ask somebody or figure it out. And I feel like that's what we can do now is go back to a bunch of the roles blurring, but like you just keep going and try to figure out what's going to work and then when you hit a wall, you have all these different sorts of scaffolding.

所以我觉得这部分没变。嗯，我觉得变了的是，这几乎像是把时钟往回拨。我刚入行做产品那会儿，恐龙还在地球上呢，那时候我们不知道——我们没有定义。我们就是不知道这份工作到底是什么。所以你就是个通用的问题解决者，四处转悠，试着去解决一个问题，直到撞上一堵墙，然后你就去问人，或者自己琢磨出来。我觉得我们现在能做的就是这个：回到一堆角色边界模糊的状态，但你就是一直往前走，试着摸清楚什么能行，等你撞墙的时候，你手边有各种各样的脚手架。

[2:40] Dan

Yeah, that makes sense. Mike, anything from you?

嗯，有道理。Mike，你有什么要补充的吗？

[2:42] Mike

I think for me, I had a real like um sort of realization moment recently and so earlier beginning of the year shifted to an IC role. So, I'm mostly building. Um, and we were getting close to shipping something. And recently, uh, and one of the like PM leads works with Ammy pulled me aside and was like, "Mike, I know you love to build. This project's going well. Like, it's looking really exciting. Like, we really need APM on this project." I was like, "I don't know. Like, do we need a PM? Like, Claude's got it. Like, we've got a lot going on." And she's like, "No, we really need a PM." And it like such a reminder like once they joined like all of the things and all the glue and all of the connective tissue and all of the things that were about to get dropped if I had not done that would that would have happened. And it was like just such a cool eye opening like before and after. Um so much so that the week later I pinged her and I was like cat you were 100% right. Um [laughter] and thank you. And it was just and I think about all the things that that encompassed. was, you know, actually bringing along, we're a company that serves everybody from, you know, proumers all the way to really large enterprises. Like, have you brought all of that contingency along?

对我来说，我最近有过一个真正的、算是恍然大悟的时刻。今年年初我转成了 IC（个人贡献者，不带团队、自己动手干活）的角色，所以我现在主要在动手造东西。嗯，当时我们快要上线一个东西了。就在最近，有一位跟 Ami 共事的 PM 负责人把我拉到一边说：「Mike，我知道你爱动手做东西。这个项目进展不错，看着真的很让人兴奋。但这个项目我们真的需要一个 PM。」我说：「我不知道啊，我们需要 PM 吗？有 Claude 顶着呢。我们手上的事已经一大堆了。」她说：「不，我们真的需要一个 PM。」结果这真是个提醒：PM 一加入，所有那些事、所有的胶水、所有的连接组织，所有那些我要是没这么做就眼看要掉地上的事——那是真会掉的。这种前后对比真的特别让人大开眼界。以至于一周之后我给她发消息说：Cat，你百分之百是对的。嗯（笑），谢谢你。然后我就在想这里面包含的所有事情。比如，真正把各方都带上——我们这家公司服务的人，从专业个人用户（prosumer）一直到非常大的企业。这一整拨人，你都带上了吗？

[3:41] Mike

Right? When uh we need to go and figure out how to enable our customer success team to like talk about the changes as questions come up in real time. Has somebody thought about that? Has somebody looped in safeguards? It's obviously very very important. Are people getting kept like kept on track? All of these things. And nobody has infinite capacity even with you know the assistance of these tools. So to be able to have like call it what you will but somebody playing that role of making sure that everything is going to go well everything is going to get connected. we've kept the end user needs in mind, Tommy's point, and that the builders who are like heads down and are like, you know, in full-on cloud mode can keep doing that without having to then pop out and do the very different activity, which is how do you actually make sure all these things go well is like I think very essential. And I think all that's happened now is that because we can move faster, that role is like you have to be operationally excellent in that role in a way that like even beyond what you had to be before. But it definitely has not gone away. In fact, I have felt this like renewed like yes, this is a like if you're recruiting for a team and you need people to wear these hats, this hat is increasingly important.

对吧？当我们需要去想办法，让客户成功团队在问题实时冒出来的时候能把这些变化讲清楚——有人想过这个吗？有人把 Safeguards（Anthropic 的安全防护团队）拉进来了吗？这显然非常、非常重要。大家是不是都还——都还在正轨上？所有这些事。而且没有人有无限的精力，哪怕有这些工具帮忙也一样。所以能有这么一个人——随你怎么叫——来扮演这个角色：确保一切都会顺利，一切都会接得上，像 Ami 说的那样我们始终把终端用户的需求放在心上，而且让那些埋头苦干、完全进入 Claude 模式的 builder（亲手把东西做出来的人）能继续干下去，不用再跳出来去做另一件很不一样的事——也就是到底怎么确保所有这些事都顺利——我觉得这非常关键。而我觉得现在发生的全部变化就是：因为我们能跑得更快了，这个角色就要求你在运营上做到卓越，甚至超出你以前需要做到的程度。但它绝对没有消失。事实上，我有一种重新被点醒的感觉：对，这是个——如果你在给一个团队招人，需要有人来戴这几顶帽子，那这一顶帽子正变得越来越重要。

[4:38] Dan

Can I ask you a serious question though?

不过我能问你一个严肃的问题吗？

[4:41] Dan

So,

那么，

[4:44]

only serious questions. Serious. This is serious.

Ami： 只问严肃的问题。

Dan： 严肃。这是严肃的。

[4:48] Dan

Why didn't you just have Fable do it? [laughter]

你当时怎么不直接让 Fable 来干？（笑）

[4:50]

Like, what is And I presumably there's some like Fable 7 you've got access to that we don't have access to yet. So, like what is that person doing that you couldn't just have Fable do? I mean it is interesting like the we have as you can imagine we like run on cloud tag and we have a lot of like you know cloud provided connective tissue too and it does catch things where you say hey like there's another thing going on in this other part of the organization that you should be aware of you know like we found it via the slash search. So there is that sort of piece as well. I don't think claude is a convenor yet. I had this um I gave a talk a couple weeks ago on like what's the role of the human in the like post super powerful AI age and there's like a bunch of archetypes I was thinking about but one of it was like the convenor like somebody still needs to bring the clouds and the people together to like get the work done and so I think that is the role that I cla organizational pull or the scheduling ability I've yet this is going to be a moment Dan when claude schedules a meeting with me or for me where I'm like who scheduled this meeting it's like oh did I thought you two should talk chat that has not happened yet [laughter] but not because it can't I guess we just like turned on that setting, you know.

Dan： 就是说，到底是——而且我猜你们手上有个什么 Fable 7，是我们还用不上的。所以那个人到底在做什么，是你没法直接让 Fable 做的？

Mike： 我是说，这确实挺有意思的。你能想象，我们基本就是跑在 Claude Tag 上的，我们也有很多由 Claude 提供的连接组织，它也确实能抓到一些事，会跟你说：嘿，组织另一个部门正在发生另一件事，你应该知道一下，你看，我们是通过 slash search 找到的。所以也有这么一块。但我觉得 Claude 还不是一个召集人。我有这么个——几周前我做了一场演讲，讲在超强 AI 出现之后的时代里，人的角色是什么，我当时想了一堆角色原型，其中一个就是召集人：还是得有人把 Claude 们和人聚到一起，把活儿干成。所以我觉得这个角色，Claude 还没有那种组织上的号召力，或者说排日程的能力——我还没——Dan，总有一天会出现这么一个时刻：Claude 跟我约了个会，或者替我约了个会，我会想「这会是谁约的？」，它就说「哦，是我约的，我觉得你们俩该聊聊」。这事还没发生过。（笑）但不是因为它做不到，我猜只是我们还没把那个设置打开而已，你懂的。

[5:51]

I [laughter]

我……（笑）

[5:54] Dan

stay tuned.

敬请期待。

[5:57] Ami

I um I think we're also living in a world where like the tools are really good and you can build a lot of things and you can build them really fast. In some ways that just like expands the universe of what is possible to build and you still need to have really good judgment about like what to build based on sometimes pretty limited information cuz it feels like everything's moving really fast. So you have to like figure out what to build and then I think you in some ways you have to be even more relentless about making sure it happens because there's so many forking paths that you could take. Yeah.

我，嗯，我觉得我们还活在这样一个世界里：工具真的很好用，你能造很多东西，而且能造得很快。从某种意义上说，这只是把「有可能造出来的东西」的范围扩大了，而你仍然需要有非常好的判断力来决定造什么，有时候依据的还是相当有限的信息，因为感觉一切都变得特别快。所以你得想清楚要造什么，然后我觉得，从某种意义上说，你得更加锲而不舍地确保它真的做成，因为你可以走的岔路实在太多了。对。

[6:36] Ami

I have an idea of what we need to do and I have the feedback loop to the people who are going to use it and I know what the human problems are and I'm just going to go for it.

我对我们需要做什么有个想法，我跟将来要用它的人之间有反馈回路，我知道人的问题是什么，那我就放手去干。

[6:43] Ami

And I think that sort of relentlessness and like judgment in the face of ambiguity is is really important.

我觉得这种锲而不舍，还有面对模糊不清时的判断力，真的非常重要。

[6:49] Dan

That is definitely something that I have found, you know, every time there's a shift, every time new models come out, it's it's sort of like you're looking out on the ocean, you're like, what's beyond the horizon? And then you get to fast forward beyond the horizon and what do you see? Well, there's more stuff. There's like a whole new landscape, you know? It's not it's not like a utopia and it's not like horrible. It's like there's just new lots of new things to discover and navigate. And I think the same is true for just models in general. when we were talking Ammy in and in in planning for this conversation, one of the things that you said and you've had a long career in product management um but prior to anthropic is that you spent decades learning how to answer specific kinds of questions as a product leader that you now no longer need to answer. And so I'm curious about okay, what are the skills that you spent such a long time building that you actually don't feel like are that useful anymore? and what are the things that you're leaning on now more and then how does that how does that filter down into the team and your expectations for people that you hire?

这一点我确实深有体会。每次有一次转变、每次有新模型出来，都有点像你在眺望大海，心想：地平线那边是什么？然后你得以快进到地平线那边，你看到了什么？嗯，还有更多东西。是一整片全新的地貌，你懂吧？它不是什么乌托邦，也不是多么可怕，就是有好多好多新东西等着你去发现、去摸索着走。我觉得放到模型上，总的来说也是一样。Ami，我们之前为这场对谈做准备、聊的时候，你说过一件事——你在产品管理上有很长的职业生涯，嗯，不过是在来 Anthropic 之前——你说你花了几十年去学，作为一个产品负责人怎么回答某几类特定的问题，而这些问题你现在已经不需要再回答了。所以我很好奇：好，哪些技能是你花了那么长时间练出来、现在其实觉得没那么有用了的？哪些是你现在更倚重的？然后这又是怎么——怎么传导到团队里，传导到你对招进来的人的期望上的？

[7:49] Ami

I think it's really true like I spent a really long time trying to understand how to put myself in the mindset of a user and how something would fit into their lives and like what we should build because it was so expensive to build and ship something. So, you know, literally this would happen where people would have reviews and what we'd talk about is like, am I where should we put this button, you know, and we would talk about the interaction pattern and where like someone's finger would go on their phone and what situation they they would be in where they pulled their phone out of their pocket and what they wanted. And I feel like I tried really hard to get good at that at that mix of like product and design. And I just like I don't think anyone's ever going to ask me that question again.

我觉得确实是这样。我花了非常长的时间去琢磨，怎么把自己放进用户的心态里，一个东西会怎么融入他们的生活，我们到底该做什么——因为做出一个东西并把它发布出去，成本太高了。所以，你知道，真的会有这种事：大家开评审会，我们讨论的是，这个按钮该放哪儿，你知道吧，我们会讨论交互模式，讨论人的手指会落在手机的哪个位置，他们是在什么情境下把手机从口袋里掏出来的，他们想要什么。我觉得我当时非常努力地想把这个练好，就是产品和设计掺在一起的那种能力。而我就是——我觉得再也不会有人问我这个问题了。

[8:29] Dan

That's a fable. That's a fable one. [laughter]

这题归 Fable。这是道 Fable 的题。（笑）

[8:30] Ami

That's in the past. Like now it's actually faster to just build three versions and try them out and be like, "Oh yeah, this works for this one." So a thing that I prided myself on is like not actually that useful anymore. And I do think that's really hard just in terms of identity. We talk about the innovator's dilemma when it comes to companies where like you're so good at one thing, you keep on trying to do it even when there's this other thing that maybe you should try doing, but you can't because you're already so good at one thing. And I feel like that's also true on a personal level where I sometimes feel like, oh, I know what I'm good at. I'm just going to keep on trying to do that even when there might be a different way to do it that might be more important in the future. And so that's what I just encourage everyone to think about. Like when you think about that product bridge, part of the technology changing every two months is you have to throw away what you knew about yourself in the job using that technology and just try something different. And most of the time I don't know if I'm going to be any good at it, you know, like I've never worked at a model company. Like I just this is not a place where I expertise and I I just kind of I feel like every couple of months I have to constantly be like, "All right, well, let's try this other thing and see if I can figure it out."

那已经是过去的事了。现在其实更快的做法是直接做出三个版本，拿来试一试，然后说「哦对，就这个管用」。所以我曾经引以为傲的一件事，其实已经没那么有用了。我也确实觉得，单从身份认同上讲，这真的很难。我们讲到公司的时候会说创新者的窘境：你在一件事上太擅长了，于是就一直想接着做它，哪怕旁边有另一件也许你该去试试的事，可你做不到，因为你已经在那一件事上太擅长了。我觉得放到个人层面上也是这样，我有时候会觉得：哦，我知道自己擅长什么，我就接着干这个吧——哪怕可能有另一种做法，在将来也许更重要。所以这就是我鼓励每个人都去想一想的。当你想到那座产品之桥的时候，技术每两个月变一次，这其中的一部分就是：你得把你对「用这项技术干这份工作的自己」的认识扔掉，直接去试点不一样的。而大多数时候我并不知道自己能不能干好，你知道吧，我以前从没在模型公司工作过。我就是——这不是我有专长的地方，我就有点——我感觉每隔几个月我就得不断跟自己说：「行吧，那就试试这另一件事，看我能不能弄明白。」

[9:42] Ami

You know,

你知道，

[9:43] Dan

what are the characteristics of someone that is able to do that well? And what have you had to develop in order to do that well?

能把这件事做好的人，有哪些特质？而你自己为了把它做好，又不得不培养出哪些东西？

[9:48] Ami

I mean, I think it comes back to the the sorts of things we were talking about. I think like raising your ceiling on tolerance for change and being willing to embrace change is super important. Like so I think adaptability I mean I'm sure this is not new. Like adaptability is just such a key thing given how rapidly the world is changing. Um, I think that sense of like judgment and relentlessness as we talked about, those are like the things that I keep coming back to and really what I try to think about for the team is like I think it's important to talk about those things because otherwise it can feel really lonely like you're the one who has to constantly adapt but actually everybody has to constantly adapt and I think it's hard on all of us. But I also talk about trying to frame the chaos so it doesn't feel as taxing,

我觉得这还是回到我们刚才聊的那些东西上。我觉得，把自己对变化的承受上限往上抬一抬、愿意去拥抱变化，这一点特别重要。所以说适应力——我知道这肯定不是什么新鲜话——世界变得这么快，适应力就是太关键的一样东西。嗯，还有我们刚才聊到的判断力和那股锲而不舍的劲儿，这些是我反复会回到的东西。而我真正替团队琢磨的是：我觉得把这些事摆到台面上讲很重要，不然人会觉得特别孤独，好像只有你一个人得不停地适应，可其实每个人都得不停地适应，而且我觉得这对我们所有人来说都不容易。不过我也常讲，要想办法给这团混乱加个框，这样它就不会那么耗人，

[10:38]

you know, like I think I think when there's times of uncertainty sometimes it's tempting to like try to map out here's exactly what's going to happen. Like here's exactly the products we're going to build. Here's exactly how my career is going to go, you know, like and that's a way to exert control. And I just I I think that locks you out of trying all the new stuff. And so what we have to do is like frame the chaos so it feels safe and it feels plausible to engage with it. But that's where like a bunch of the magic and productivity is going to be. So just like talking about it, making it safe, acknowledging that there's also emotions around all of this and it's not easy, but it is a way to just participate and understand and work with the future. I think that's something we talk about a lot. I think there's an organizational leadership question that I think Amy you do really well which is like you can have the chaos but you also need very clear leadership and DRI ship within some areas. So like you can have a bunch of people sort of exploring something very frontier but having a clear like this is the person who's ultimately going to make the call whether this is you know the thing to pursue or not. In labs we call them leads or like bet we have call them bets within labs. So the bet lead is a really important role and they are the DRRi for saying like I think we should double down on this or like actually we should wind this down. this team needs more people, this team needs less. And like that role is even more important because we're moving so quickly, because things are shifting really rapidly to still have that like we're all lost together, but somebody's got the like, you know, the pen on like what what's the next most important thing we can do to derisk, to understand, to learn, to move things forward, too.

Ami： 你知道，我觉得在充满不确定的时候，人有时候会忍不住想把一切都规划出来：接下来到底会发生什么，我们到底要做哪些产品，我的职业生涯到底会怎么走，你懂的，这是一种把控制权抓在手里的办法。可我就是觉得，这么干会把你挡在所有新东西的门外，试都没法试。所以我们要做的，是给混乱加个框，让人觉得去碰它是安全的、是可行的。而一大堆神奇的东西和生产力恰恰就在那儿。所以就是把它拿出来聊，让它变得安全，承认这一切背后也带着情绪、并不容易，但这确实是一条能让你参与进来、去理解未来、和未来共事的路。我觉得这是我们经常聊的一件事。

Mike： 我觉得这里面还有一个组织领导力的问题，Ami，我觉得你在这上面做得特别好：混乱可以有，但在某些领域里，你也需要非常清晰的领导和 DRI（直接负责人）机制。就是说，你可以让一帮人去探索某个非常前沿的东西，但得有一个明确的人，由他最终拍板这东西到底值不值得追下去。在 Labs，我们管这叫 lead，或者说——Labs 里面我们把这些项目叫 bet（押注）。所以 bet lead 是个非常重要的角色，他们就是那个 DRI，负责说「我觉得这个我们应该加倍下注」，或者「其实这个我们该收掉了」，这个团队需要加人，那个团队需要减人。而且正因为我们跑得这么快、事情变得这么急，这个角色反而更重要了：得保住那种状态——我们大家是一起迷路的，但总有个人，你知道，握着笔，来定下一步最重要的事是什么，做了它我们就能降低风险、弄明白状况、学到东西，也把事情往前推。

[12:09] Dan

We've talked a lot uh already so far about uh the psychology and the skills of being a product leader. Um, one other place that's changing a lot is what you actually build and how you think about like the features and the functionality in particular because we're moving from a world where you'd expect only humans to use your software to a world where uh humans and agents are using software. Sometimes it's delegated, sometimes it's fully collaborative. And I know Mike, you've sort of been at the center of figuring out what is what does agentnative type software look like? And I think you have a lot of thoughts on where that is going. So, can you talk to us about the uh h how you think about what to build, what goes into a product versus an agent and uh and what's what's the overlap?

到目前为止，我们已经聊了很多当产品负责人的心理和技能。嗯，还有一个变化很大的地方，是你到底做什么，以及你怎么去想功能和特性这些东西，尤其是因为我们正从一个你只指望人来用你软件的世界，走向一个人和 agent 都在用软件的世界。有时候是把活儿委派出去，有时候是完全协作。我知道，Mike，你差不多一直就在这件事的中心，在摸索 agent 原生的软件到底长什么样。而且我觉得你对它往哪儿走有很多想法。所以，能不能跟我们讲讲，你怎么想该做什么、什么该放进产品里、什么该交给 agent，以及，两者重叠的部分是什么？

[12:56] Mike

I love this question and like I maybe there's like a timeline to look at where V1 or like er 1 was great, we've got this AI thing, maybe it belongs in like a sidebar or like a little mini window and you can ask questions and it's kind of disconnected and if you know maybe it helps you with some customer support questions. Um and then sort of greater integrations where maybe there are whole features that are AI powered and then this movement towards being more agent native and I think every like literally wrote the book on it which is and the way I for that the whole year where I was working a lot in this idea of like agent native architectures I would basically feed their like write up on this and then create a skill out of it because I was really good encapsulation of it was the idea of like everything a human could do an agent should be able to do as well and very few products actually get this right. I think a lot of our products like don't quite get this right yet. But when you do that, all of a sudden sort of emergent behavior gets unlocked where you know the agent can sort of piece together things that weren't pieced together before or can proactively offer um different ways of doing things. I think we're still on the journey of making that good, but I think the next stage that I've been thinking about a lot is when the interfaces actually become something that is malleable by the agent too. And we've talked about mal malleable software for years. I feel like it's actually now like really manifesting and becoming real. And um maybe back to the sort of role of product and connecting to this question, one of the things that we have, we have like a pretty complex thing that we're trying to ship right now. And it's got at least four independent orc streams.

我很喜欢这个问题。我觉得也许可以拉一条时间线来看：V1，或者说第一个时代，是「太好了，我们有了 AI 这个东西，也许它该放在侧边栏或者一个小小的迷你窗口里，你可以问它问题」，它跟别的东西有点脱节，也许能帮你解决一些客服问题。嗯，然后是更深的集成，可能有一整块一整块的功能是 AI 驱动的；再然后就是朝着更 agent 原生的方向走。我觉得 Every 真的是把这件事写成了教科书。那一整年我都在大量琢磨 agent 原生架构这个想法，我的做法基本上就是把他们写的那篇东西喂进去，再据此做成一个 skill，因为它概括得真的很好。那个想法就是：凡是人能做的，agent 也应该都能做。而真正把这一点做对的产品非常少。我觉得我们自己很多产品也还没完全做对。可一旦你做到了，一下子就会解锁某种涌现行为：agent 能把以前没拼到一起的东西拼起来，或者能主动提出不一样的做法。我觉得把这件事做好，我们还在路上；但我一直在想很多的下一个阶段，是界面本身也变成 agent 可以塑造的东西。可塑软件我们已经聊了好多年了，我感觉它现在才算真正在显形、在变成现实。嗯，也许回到产品的角色上、接上这个问题：我们手上有一件——我们现在正想上线一个相当复杂的东西，它至少有四条相互独立的工作线。

[14:17] Mike

And the way we've operationalized it is we've had Claude both sort of keep an eye on how everything is going, but also create the UI by which first the TPM and then the entire team understands what is happening in that project. And what's really cool about that is that if you didn't like that particular display, you're not bound to like however it was determined to be built either by some internal acceleration team or some uh sort of third party. You can change it um and you can change it along with cloud and it's just been a very cool thing to see internally. Like one of the biggest shifts in how we've worked in the last year I think at Enthropic is so much of the software we interact with is like built maintained and iterated on by claude. of course reacting to all the things that are happening in the organization but very much like changing all the time and so that leads to all sorts of other interesting questions like who can update the data right is it just the humans clicking through is it also clawed in the background how do you have like provenence of data it opens up some like interesting questions there as well but I feel like that is the piece like the both the trend and the thing I'm most inspired by right now is how do we make that the way almost all of our software operates and of course you still want building blocks and predictability and design systems and you don't want you want to frame the cats you want to like make it not absolutely insane, but for the given task, for that given project, for even that given user potentially, what is the way in which we pull in the right um ways and make that really feel both personal and like extremely useful because of that

我们把它落到实处的办法是：让 Claude 一边盯着各条线的进展，一边把界面也做出来——先是 TPM（技术项目经理）、然后是整个团队，都靠这个界面来了解项目里正在发生什么。这里面特别酷的一点是，如果你不喜欢那种展示方式，你并不会被绑死在它当初被定下来的做法上——不管那是某个内部的提效团队定的，还是某个，呃，第三方定的。你可以改它，嗯，而且可以跟 Claude 一起改，在内部看到这个真的很酷。我觉得过去一年 Anthropic 工作方式上最大的变化之一，就是我们打交道的软件里有那么多是由 Claude 搭建、维护、迭代的。它当然是在回应组织里发生的各种事，但确实是一直在变。这就引出了各种别的有意思的问题，比如谁能更新数据，对吧？只有人点来点去才行，还是 Claude 也在后台更新？数据的来源出处你怎么追溯？这也打开了一些挺有意思的问题。但我觉得这就是那一块——既是趋势，也是眼下最让我受启发的东西：怎么让我们几乎所有的软件都这样运转。当然，你还是想要积木块、可预期性和设计系统，你不想——你想给混乱加个框，想让它不至于彻底乱套；但针对那个具体的任务、那个具体的项目，甚至可能是那个具体的用户，我们该用什么方式把合适的，嗯，做法拉进来，让它真的既有个人感、又因此格外好用

[15:34]

and how does someone think about maybe you know you're not in anthropic and you're you're have you have a SAS app that's scaled and you're thinking about okay, do I put an agent in it? Do I make it open and agent native so that cloud can be in it? like how do how do you think about those choices and what a company should care about versus uh leave to leave to a model company like Anthropic to to plug in? Yeah, I think a lot of it is really building the right primitives into the product that almost like an infrastructural layer and you can kind of tell like I I always poke at the whenever I'm using new products um I'll ask it to do something that you know a human could do but you know might be required somebody to think through or architect the thing where everything goes through some shared plumbing that like both agents and like whatever sort of rest API you've built can go through as well and you can tell when they've been thought through that way and it's I have a lot of empathy because if you've been building an application for 20 years just bolting that on is difficult. So I think is a journey that a lot of companies are going on. But once you get it right, I think that question becomes you can just evolve it over time, right?

Dan： 那换个人该怎么想这件事呢——比如说，你不在 Anthropic，你手上有一个已经做出规模的 SaaS 应用，你在想：好，我要不要往里面放一个 agent？我要不要把它做成开放的、agent 原生的，好让 Claude 能进来？这些选择你怎么看？哪些是一家公司自己该操心的，哪些，呃，该留给 Anthropic 这样的模型公司来接入？

Mike： 对，我觉得很大程度上其实是把合适的基础原语（primitives，可复用的底层积木）做进产品里，差不多像一层基础设施。而且你大概能看得出来——我每次用新产品都喜欢戳一戳，嗯，我会让它去做一件人能做、但可能得有人事先想清楚、把架构搭好的事：所有操作都走同一套共享管道，agent 也好、你做的不管哪种 REST API 也好，都从那儿过。它们是不是照这个思路想过，你是看得出来的。我其实很能体谅，因为如果一个应用你已经做了 20 年，硬把这个东西外挂上去是很难的。所以我觉得这是很多公司正在走的一段路。不过一旦做对了，我觉得那个问题就变成：你可以随着时间慢慢演进它，对吧？

[16:31] Mike

You can start with, okay, at first I'm just going to have maybe more of the side paint thing. I don't I'm not ready to blow up my whole UI. I'm not ready to create fully generative UIs, but maybe you start experimenting with like, all right, there's also maybe a personal landing page that can be more malleable and is using the same primitives. You're not inventing a whole new sort of infrastructure underneath. So I think that you can evolve your product along the way, but you have like you can tell when things feel super bolted on and like like that's you know feeling like it's not a native part of the product.

你可以这样起步：好，一开始我也许就先多做点侧边栏那类东西。我不——我还没准备好把整个界面推倒重来，也没准备好做完全生成式的界面，但你也许可以开始试一试，比如：行，也许还可以有一个更可塑的个人落地页，用的是同一套基础原语。你并没有在底下另造一整套新的基础设施。所以我觉得你可以一路演进你的产品，不过你会——东西要是硬生生外挂上去的，你是看得出来的，就是那种，你知道，感觉它不是产品原生的一部分。

[16:57] Dan

So Ammy I'm curious we're hearing we can go all the way to you know agent native malleable software but uh anthropic you have to serve this whole spectrum of people. there's this uh challenge of pushing the frontier and also you know some people just want to use cloud to like chat or you've you you have big enterprise contracts where you can't just like rip out their whole interface and then say hey it's malleable like you know on on in one day or maybe you can't I don't know but I assume you can't so how do you balance

那 Ami，我很好奇——我们刚听到的是，可以一路走到，你知道，agent 原生的可塑软件，但是，呃，在 Anthropic，你们得服务这一整条光谱上的人。这里有个，呃，难题：既要推进前沿，同时，你知道，有些人就只想用 Claude 聊聊天，或者你们签着大型企业合同，没法直接把人家整个界面拆掉，然后说「嘿，现在它可塑了」，你知道，一天之内就这么干——也可能你们可以，我不知道，但我猜是不行的。所以你怎么平衡

[17:29]

[snorts]

（笑）

[17:29] Dan

uh in terms of thinking about product pushing the frontier and creating a a consistent experience for users that they can understand and bringing along the people who are maybe not going to try all the new things on day one.

呃，在考虑产品的时候：一边推进前沿，一边给用户做出一个他们能理解的、连贯一致的体验，同时还把那些可能不会在第一天就去试所有新东西的人一起带上。

[17:42]

I think it's a really hard problem. Um I mean our approach is to just try to meet everyone where they are. I think we're just so early in like understanding what works. It's not like stuff is stable and we can just build the right answer. It's like we're all trying to explore together. And part of the everything changing every couple months is like the models change, but also like the way people use products and the models changes and we kind of constantly have to update for that. And so a thing that I think about is like when we're just exploring a a product like like Mike's talking about, it's okay to just try stuff like we just have to try a bunch of things because we haven't actually explored the frontier. And then when we figure out what works, then we have to like make it make sense for all the people who want something that feels more stable and complete. Um, I think this is a really tough trade-off because, you know, like raised in like the the product like world that I was, I was all about like simplicity and reliability and total predictability because I think that helps people build like a muscle memory and an intuition. But I also think the reality is like we just don't know yet what a lot of things should look like. And so I actually encourage the team to just like take a lot of shots on goal. You know, it's okay to have things that feel somewhat overlapping or duplicative as long as when we figure them out. When we hit product market fit, we go back and we reinccorporate into something that works for everyone. Because I feel like the fail state is you decide from the beginning like based on a theory of exactly what needs to happen and you like over constrain a product. And I would much rather have like five overlapping products that all work in different ways than like one overly constrained one that we never even gave a chance to actually like meet the market. And how do you and I definitely

Ami： 我觉得这是个非常难的问题。嗯，我们的做法就是尽量在每个人所在的地方接住他们。我觉得我们在「弄明白什么行得通」这件事上还太早了。并不是说东西已经稳定了、我们把正确答案做出来就行，而是我们大家都在一起探索。所谓每隔几个月一切都在变，一部分是模型在变，但人们使用产品和模型的方式也在变，我们差不多得不停地为此更新。所以我会想的一点是：当我们只是在探索一个产品的时候，就像 Mike 说的那样，随便试是没问题的，我们就是得试一大堆东西，因为前沿我们其实还没探索过。然后等我们搞清楚什么行得通，就得把它整理得说得通，给所有想要更稳定、更完整的东西的人用。嗯，我觉得这是个很难的取舍，因为，你知道，我是在那样一个产品世界里长大的，我以前讲究的全是简洁、可靠、完全可预期，因为我觉得这能帮人养成肌肉记忆和直觉。但我也觉得现实是，很多东西该长什么样，我们现在就是还不知道。所以我其实鼓励团队多起脚射门。你知道，有些东西感觉有点重叠、有点重复也没关系，只要等我们把它们搞明白了——等我们打中了产品市场契合（PMF），再回过头把它们重新收进一个对所有人都好用的东西里。因为我觉得失败的状态是：你从一开始就根据一套「到底必须怎么发生」的理论拍了板，结果把产品约束得太死。我宁可有五个相互重叠、各有各做法的产品，也不要一个约束过头、我们连让它真正去见市场的机会都没给过的产品。

Dan： 那你们怎么——而且我确实

[19:32] Dan

see this we we do this at every but like I see this all around the industry just doing lots of parallel experiments and seeing which ones one one works and then starting to consolidate as you start to get uh get signal. How do you do that in a way that doesn't feel confusing or incoherent or you know if I'm working on the third team doing this thing like it's can be maybe demotivating like how do you do that? Well

看到这种做法——我们在 Every 就这么干，不过我看整个行业到处都是这样：同时跑很多并行的实验，看哪一个行得通，然后等开始，呃，拿到信号了再着手收拢。你们怎么做到让这件事不显得混乱、不显得前后不连贯？或者，你知道，如果我在第三个做同一件事的团队里，那可能挺打击士气的。所以这事你们是怎么做的？嗯，

[19:54] Mike

I think there's a couple of things. one like I think when we were just having this conversation this morning internally where you know there was like two potential like product directions and one has like a team with very high conviction going after and the other one doesn't really and like it feels very manufactured to be like we should explore this other idea cuz we think it's good but nobody's like really calling for it then that's like that kind of B team feeling that I think I've never been like successful tasking a team to go like take a product area that they're not excited about like the product is going to suck at the end of that process right um so like you want you want the like excitement but if there are two three different teams that have exciting directions and yeah like let that let that explore. I think I I think we've made a lot of progress on even in the last six months is like having the underlying infrastructure to support all of that. So, we inventoried this. It feels like 6 months ago. It's probably only three, but like there was a point where like chat and co-work had different memory systems and different MCP and different um ways of storing files. And it was like, okay, like for anybody to experiment on top of the surface, they're going to have a really hard time because if they built a third thing that then doesn't share memory with any of those, for example, it's going to feel disconnected and like almost disadvantaged from the beginning.

我觉得有几点。第一，我们今天早上内部刚好在聊这个：当时有两个可能的产品方向，一个有一支信念特别强的团队在追，另一个其实没有。这时候要是硬说“我们也该探索一下另外那个想法，因为我们觉得它不错”，可又没有谁真的在喊着要做，就会显得特别刻意，那就是那种“B 队”的感觉。我从来没成功过的一件事，就是指派一支团队去接一个他们自己不兴奋的产品领域——这么走一遍下来，产品肯定很烂，对吧。所以你要的是那股兴奋劲儿。但如果有两三支团队各自都有让人兴奋的方向，那行，就放手让它们去探索。另外我觉得，我们进步很大的一点——哪怕就在过去这六个月里——是有了能支撑这一切的底层基础设施。我们盘点过一遍，感觉像是 6 个月前的事，其实可能才三个月。当时有那么一个阶段，chat 和 Cowork 用的是不同的记忆系统、不同的 MCP、不同的文件存储方式。那局面就是：谁想在这些产品界面之上做实验，都会特别费劲。比如说，他们要是做出了第三个东西，它又跟前面哪个都不共享记忆，那它用起来就是割裂的，几乎从一开始就处在劣势。

[20:59] Mike

So, a lot of the work, you know, we have like a foundations team that has like been just thinking about like you should be able to get your memory wherever you are within our products. And like it sounds like a simple thing but then it's very clearly you pull back the thing like oh it's complicated for these eight reasons but it's worth solving that and solving it in a different team than like your labsy innovation or like kind of frontier product teams because then like that those experiments can feel complimentary even if they are overlapping too.

所以很多工作是这样的——我们有一支 foundations（基础）团队，一直在琢磨的就是：不管你在我们产品的哪个地方，都应该能拿到你的记忆。这听着是件挺简单的事，可你把盖子一掀开就很清楚了：哦，原来因为这八个原因，它很复杂。但这件事值得解决，而且值得放在另一支团队里去解决，而不是交给那些偏 Labs 式创新、或者说做前沿产品的团队。因为这样一来，那些实验哪怕彼此有重叠，也能让人感觉是互补的。

[21:20] Ami

I also think like this sounds so simple but I also think it's just really important to talk about it and just name like hey if there's three different bets sometimes it can feel demoralizing or like one of you is not set up for success. That's okay. That's the world we're living in. And so what we're asking you to do is come along this journey with us. We don't know the answer. We don't think anyone knows the answer. We're all going to discover it. And we're all fundamentally on the same team. And so we will just like learn from whatever works. And just like saying stuff like that out loud, I think is really important because everybody is like trying to figure everything out in the dark.

我还觉得——这话听着特别简单——把这件事拿出来讲、直接点破，真的很重要。比如说：嘿，如果同时有三个不同的赌注，有时候确实会让人泄气，或者感觉你们当中有一个从一开始就没被放在能成功的位置上。没关系，我们就活在这样一个世界里。所以我们想请你做的，是跟我们一起走这段路。我们不知道答案，我们也不觉得有谁知道答案，我们会一起把它找出来。而且说到底，我们都在同一支队伍里。所以不管最后是哪个跑通了，我们都从里面学东西。我觉得光是把这类话大声说出来，就非常重要，因为每个人都是在黑暗里摸索着想把一切弄明白。

[21:55] Dan

What do you look for to know that something's starting to work?

你们看什么信号，来判断一个东西开始跑通了？

[21:58] Ami

I mean, I think a bunch of that is kind of the normal product market fits. Like do people use it? Do they like it? Do they come back? Do they talk about it? Do they get value? Like again going back to like the goal of product has remained the same. It's like to solve problems. Thinking about does this actually solve a problem in a way that people can use and use easily and feel good about using it comes back to the basics for that.

我觉得其中很大一部分，其实就是常规的那套产品市场契合（PMF）：大家用不用？喜不喜欢？会不会回来？会不会跟别人聊起它？有没有从中得到价值？还是回到那句话——产品的目标一直没变，就是解决问题。去想一想：它是不是真的解决了一个问题，而且是以一种大家用得上、用得轻松、用着感觉也好的方式？这件事说到底还是回归基本功。

[22:19]

One of the interesting things that feels challenging to me about that is sometimes a thing looks like it doesn't work and the new model comes out and then it really starts working and you don't want to like throw it away too early but you also don't want to be too far ahead. So how do you do that? Like one of the things we do in labs sometimes is just park a project. Like when we built we built the first like computer use product inside Anthropic in 2024. Um and it was so bad because the models just weren't there yet. And you would be you know our first thesis was well it's not going to be good at a bunch of automation but it might be good about education. Um, so you'd be like, "All right, uh, how do I, you know, change this, you know, thing in some complex software? Like, how do I, you know, do this really complex thing in Photoshop?" And it would get there, but it would get there in like the very like early agent cloud way, which was like this, did that work? Nope, that didn't work. I got to close this. It was like 20 minutes later like, I did it. You're like, I learned nothing. I learn, you know, so like it was neither good for automation nor education. But we parked it and then we would basically just try it with every new model. And we actually, the way we would try it is we actually just had it like running in an eval harness. And the way we actually learned that 37 was like a like a leap in computer use was like just seeing it suddenly took over and then we actually looked at the transcript. We're like wow it's succeeding more often than it's not like this is a real moment. So it is really valuable like we see it as a win in labs if what you did was all if all your project did is show that the models aren't ready for it yet but you're able to now like externalize that as an eval or sort of like create some uh connectivity to the research team like that's great because it means that 2 3 4

Dan： 这里面有一点很有意思，也是我自己觉得挺难的：有时候一个东西看着不行，结果新模型一出来，它就真的好用起来了。你不想太早把它扔掉，可也不想超前太多。那该怎么办呢？

Mike： 我们在 Labs 有时候会做的一件事，就是把项目先搁着。比如我们在 2024 年做出了 Anthropic 内部第一个 computer use（让模型直接操作电脑）产品，它特别烂，因为那时候模型根本还没到那个水平。我们最初的设想是：好吧，它做各种自动化不会好用，但没准适合用来教学。所以你会问它：“好，我要怎么在某个复杂软件里改这个东西？比如这个特别复杂的操作，在 Photoshop 里要怎么做？”它是能做到的，但做到的方式是那种很早期的 agent 式 Claude 的路数，就是：试这个，那个成了吗？没，没成。我得把这个关掉。大概 20 分钟以后它说：我做完了。而你的反应是：我什么也没学到。所以它既不适合自动化，也不适合教学。但我们把它搁在了那儿，然后基本上每出一个新模型就拿出来试一遍。我们试的方式，其实就是让它一直跑在一个 eval harness（评测框架）里。我们之所以知道 3.7 在 computer use 上是一次飞跃，就是因为眼看着它突然一下子起来了，然后我们去翻了 transcript（运行记录），心想：哇，它成功的次数已经比失败的多了，这是个实打实的转折点。所以这真的很有价值——在 Labs，如果你的项目做到的全部事情，只是证明了模型还没准备好，但你现在能把这一点外化成一个 eval，或者说跟研究团队建立起某种连接，我们就把它算作一次胜利。这很好，因为这意味着再过两、三、四、

[23:43] Mike

5 6 months from now we'll be able to revisit and hopefully we've learned something along the way um as well but I think you need you need to build early so that you can even have that intuition.

五、六个月，我们就能回头再看它，而且但愿这一路上我们也学到了点东西。不过我觉得，你得早点动手做，才有可能拥有那种直觉。

[23:51] Dan

I definitely agree. I've been calling this like it's it's sort there's a sort of you can get capability blind if you try something and then you don't ever try it again or you don't make an eval and you're like oh the models are just never going to do this and then it just turns out a model comes out like in 3 months and it just completely chang it it improves in ways you never would have thought it would. I'm I'm actually curious if you're here how many of you have used Fable? Raise your hand. Oh, a lot of people are using Okay, this is great. This is a good What about Astra? Fewer. Interesting. Okay, so this is great. you guys are you guys are on the you guys are on the edge. This is this is very helpful because one of the things I find a lot is I talk to someone and they're like, "Oh, it doesn't do this." And I'm like, "Have you used the latest model?" And they're like, "No, I used like Sonnet 45 or what?" Like whatever. And uh that that really changes things.

我非常同意。我一直管这个叫——就是有这么一种情况：你可能会陷入“能力盲区”（capability blind）。你试过一样东西，之后就再也不试了，或者也没做个 eval，然后你就认定：哦，模型永远干不了这个。结果过了 3 个月出了个新模型，它就彻底变了——它进步的方式是你怎么也想不到的。其实我挺好奇的，在场的各位，有多少人用过 Fable？举个手。哦，好多人都在用。好，这很棒，这是个好……那 Astra 呢？少一些。有意思。好，这很棒，你们都走在最前沿。这一点很有帮助，因为我经常碰到这种事：我跟人聊，对方说“哦，它干不了这个”，我就问“你用过最新的模型吗？”对方说“没有，我用的是 Sonnet 4.5 还是什么的”，反正就那类。而这一点真的会让结果很不一样。

[24:36] Ami

I think that's also part of why it's so important to be willing to throw away everything, you know.

我觉得这也是为什么“愿意把一切都扔掉”这么重要，你懂的。

[24:40]

Yeah.

对。

[24:40]

Even though it's hard because then otherwise you just like think that the last thing that happened is going to be true forever. One one big question that I I see you guys working on from afar and I see this happening at OpenAI 2 and and we run into it a little bit is when you're doing all these parallel experiments and one starts to really work then you have to like push it back into your actual product and there are several different ways to do it. It's like okay you can make a tab but then you have a bunch of different tabs and there's all all the things like only work in like in slightly different ways and that's kind of ugly. Or you can have it just be a separate app and you never merge them. There's all these different things that I don't think anyone has really come to this is the answer, this is how you do it. But it seems like a critical problem is how to take those parallel experiments and turn it into a coherent product. How have you what have you learned and how do you think about it?

Ami： 虽然这很难。因为不这样的话，你就会以为上一次发生的事会永远成立。

Dan： 有一个大问题，我从远处看到你们在解决，也看到 OpenAI 那边在发生，我们自己也多少碰到过：当你同时跑着这么多并行实验，其中一个真的开始跑通了，你就得把它推回到你真正的产品里去，而做法有好几种。比如，你可以做成一个 tab，可这样你就有了一堆不同的 tab，里面各样东西的运作方式都只是略有不同，挺难看的。或者你可以就让它当一个独立的 app，永远不合并。做法五花八门，我觉得还没有谁真正得出“这就是答案、就该这么做”的结论。但这看起来是个很关键的问题：怎么把那些并行实验变成一个连贯的产品。你们学到了什么？又是怎么思考这件事的？

[25:29] Ami

I don't know what you're talking about. [laughter] Um, you're right. I don't think we've fully figured it out. But I think some of it comes back to what Mike was talking about of making sure that we're building primitives so that our products feel like the way they should feel, you know, when you're using them. Like what does a user expect? The user expects to walk up to something and like hopefully not have to think about it too much. Like and right now it's easy to just transfer a bunch of cognitive load to the user and be like here's some tools like why don't you choose? And so one of the things we that that Mike's talking about is like how do we make sure that we build primitives so the right system knows you. So a UI that feels like really accessible all the time. And I think all of that is a bit iterative as we see how people react.

我不知道你在说什么。（笑）嗯，你说得对，我觉得我们还没完全想明白。但我觉得其中一部分又回到 Mike 刚才讲的：要确保我们在构建基础原语（primitives），这样我们的产品用起来，才是它们该有的那种感觉。用户期待的是什么？用户期待的是走到一个东西跟前，最好不用想太多。而现在很容易就把一大堆认知负担转嫁给用户，说：“这儿有几个工具，你自己挑吧？”所以 Mike 讲的其中一点就是：我们怎么确保把基础原语建好，让对的那套系统了解你，于是界面让人觉得随时都触手可及。我觉得这一切都带点迭代的性质，要边看大家的反应边调整。

[26:17] Mike

I think overall like if I had to bet like there's some lessons you'll learn and some that are are real. Like the thing that we always saw at Instagram it's like there's a very strong power law in tab usage. It makes sense. every single product right it's like main feed was like 80% of the use right and you can make explore better and maybe you would go from 10 to 15% but it was like main feed that is where you land that is your big thing so I think that is will always be the journey that we're all on which is like what is the main what is the main tab like what is that experience how do you still make that great and then luckily it can be an entry point to a lot of other downstream interesting experiences but yeah I think half of the job is saying no to another thing in the sidebar or another tab or another thing and figuring out Can this be rolled in once we've proven it out but not killing it so early. So this is the entire I think art of product development I think as things are moving this quickly.

我觉得总体上，如果非让我押的话，有些经验是你会学到的，有些是实打实成立的。比如我们在 Instagram 一直看到的：tab 的使用存在非常强的幂律。这很合理，每一个产品都是这样，对吧——主 feed 大概占了 80% 的使用量。你可以把 Explore 做得更好，也许能从 10% 涨到 15%，但主 feed 才是你落地的地方，那才是你的重头。所以我觉得，这会一直是我们所有人都在走的路：主 tab 是什么？那个体验是什么？你怎么还能把它做得很棒？然后幸运的是，它可以成为通往下游很多有趣体验的入口。不过对，我觉得这份工作有一半是在说不——对“侧边栏里再加一个东西”“再加一个 tab”“再加一个别的什么”说不，并且想清楚：等我们验证过了，它能不能被并进来——但又别太早把它砍掉。所以我觉得，在事情变化这么快的时候，这就是产品开发的全部艺术所在。

[27:04]

All right, we've got time for one last question which I'd love to hear from both of you which is if if we are on this stage here next year and Lenny invites us back. I don't know if he'll invite me back. I've been cursing too much but he'll invite you two back. What what do you expect will be different and and how will being a PM have changed? I hope it just like that that people feel like they can just do much more like they can just go and build in whatever way they want. They can get answers whenever they they want. They can start businesses and like and we just see a lot more like small teams and solo empowerment and we we just we start to get the uplift of like what that means for everyone else. Like I think it I think it's always like builders magnify impact for a bunch more people who will never like use the tools directly but who can benefit from people who do build things. And so I just I hope that's just way easier.

Dan： 好，我们还有时间问最后一个问题，我想听听你们两位的回答：假如明年我们还站在这个台上，Lenny 又把我们请回来——我不知道他还会不会请我，我今天脏话说得太多了，不过他肯定会请你们俩——你们觉得到时候会有什么不一样？当 PM 这件事又会变成什么样？

Ami： 我希望到时候大家会觉得自己能做的事多得多：想用什么方式去造东西，就能直接去造；想要答案，随时都能拿到；可以去创业。我们会看到多得多的小团队，看到更多单枪匹马的人被赋能，然后我们开始享受到这对其他所有人意味着的那份提升。我觉得一直都是这样：builder（亲手把东西做出来的人）会把影响力放大给更多的人——那些人永远不会直接去用这些工具，却能从真正动手造东西的人那里受益。所以我就希望，这件事能变得容易得多。

[27:56] Mike

Yeah. I think the biggest one for me is you know right now like I often talk about like the gap between what the models are capable of and like what most folks are using them for. Not through any fault of their own. It's on us. Like we got to build the products to make that possible. And like that was a hard thing last thing 24. was even harder last year. It's getting even harder now. And so like the biggest thing I want to do is close I think it's very democratizing if you can close that gap because it means it's not just your most like clawpilled software engineer that has you know like this fully working multi-ogenic setup. It's also the person that is doing like a bunch of research in parallel. It's also the person that's trying to manage their business but they feel like they've unlocked it and we've brought them along the way in terms of how do you like understand the moving parts, check in at the right places, do long horizon work but not in a way that feels disempowering. That's the art. Like if we succeed a year from now, that'll be closer.

对。对我来说最大的一点是——我现在经常讲，模型的能力和大多数人实际拿它们来做的事之间，有一道差距。这不是用户的错，责任在我们，得由我们把产品做出来，让这件事成为可能。这事在 24 年就很难，去年更难，现在是越来越难。所以我最想做的就是把这道——我觉得如果能把这道差距补上，那是一件非常普惠的事。因为这意味着，不再只有你身边最 Claude-pilled（对 Claude 最上头）的那位软件工程师，才拥有一套完全跑得起来的多 agent 配置；那个同时并行做着一堆调研的人也有，那个想把自己的生意打理好的人也有。而且他们会觉得自己把它解锁了，是我们一路带着他们走过来的：怎么看懂那些活动部件，怎么在对的节点上过问一下，怎么做长周期的工作，却又不让人觉得自己被架空了。这就是其中的艺术。如果我们成功了，一年之后，我们会离这个更近一些。

[28:43] Dan

Amazing. Thank you so much and Mike to [applause] see you.

太棒了。非常感谢 Ami 和 Mike。（掌声）很高兴见到你们。