The Expensive Part Is Taste

today my guest is Andrew Ambrosino, product and engineering lead for the Codex app at OpenAI.

今天我的嘉宾是 Andrew Ambrosino,OpenAI Codex app 的产品和工程负责人。

Codex is quickly becoming people's go-to app for building products and also for non-product work like organizing files in your computer, drafting documents, doing data analysis, reading your emails, and a lot more.

Codex 正迅速成为大家做产品的首选 app,也被用来做很多非产品类的工作,比如整理电脑里的文件、起草文档、做数据分析、读邮件等等。

If you stick around for the end of this episode, we actually have a little clip from after we stopped recording where the producer in the room started talking about how he uses Codex in his editing work.

如果你听到这期节目的最后,我们还放了一小段停止录制之后的花絮——房间里的制作人聊起了他怎么在自己的剪辑工作里用 Codex。

Since this January, Codex usage has grown 6x.

今年 1 月以来,Codex 的使用量增长了 6 倍。

They currently have over 5 million weekly active users.

他们现在的周活跃用户超过 5 million。

I suspect this number is quickly going to be out of date.

我猜这个数字很快就会过时。

Internally at OpenAI, nearly 100% of their employees use Codex weekly.

在 OpenAI 内部,几乎 100% 的员工每周都在用 Codex。

And that is not just the engineers.

而且不只是工程师。

Andrew is a designer turned engineer turned product manager who is building the app that more and more of the world is using to build their own products.

Andrew 从设计师转做工程师、再转做产品经理,如今他打造的这个 app,正被越来越多的人用来做他们自己的产品。

Before we get into it, don't forget to check out lennysproductpass.com for a year free of the hottest and most well-crafted AI products in the world available exclusively to Lenny's newsletter subscribers.

在进入正题之前,别忘了去 lennysproductpass.com 看看——那里有全世界最热门、打磨得最精良的 AI 产品,免费用一年,只对 Lenny's newsletter 的订阅者开放。

With that, I bring you Andrew Ambrosino.

那么,有请 Andrew Ambrosino。

Andrew, thank you so much for being here.

Andrew,非常感谢你来。

Welcome to the podcast.

欢迎来到播客。

Thank you for having this is a rare in-person podcast.

谢谢你请我来,这是难得的一次线下播客。

I rarely do this kind of thing.

我很少做这种事。

We'll see how it goes.

看看效果怎么样吧。

We'll see.

看看吧。

We'll see people like these more.

以后这类线下的我们会多做一些。

When we were preparing for this chat, I asked you what's the biggest thing you want people to get out of this conversation and you said that it was how AI is changing the shape of product work.

我们准备这次聊天的时候,我问你最想让听众从这次对话里得到什么,你说是 AI 正在怎样改变产品工作的形态。

You're working at maybe the most bleeding edge AI pill software team there is.

你所在的团队,大概是当下最前沿、最信 AI 那一套的软件团队。

So you have a really interesting lens into where things are heading, where other teams are going to be and in a year or two or or more.

所以你有一个很有意思的视角,能看到事情正往哪走、别的团队一两年甚至更久之后会走到哪。

What does the shape of product team look like now versus a couple years ago?

跟几年前相比,现在产品团队的形态是什么样的?

One of the the hardest things to do right now as a leader building these products is just sort of the inversion of the process in my mind which I think a lot of people have talked about which is that anybody can build anything right like I I generally believe now that starting from scratch if you talk to these models ours anybody else's really um you can stand up whatever feature you want right And that's not necessarily a hard part of software, but that's like that's really cool.

现在作为打造这类产品的负责人,最难的事情之一,就是我脑子里那个流程的反转——我想很多人都聊过——就是任何人都能造出任何东西:我现在基本上相信,哪怕从零开始,只要你去跟这些模型对话,不管是我们的还是别人的,你想要什么功能都能搭起来。这不见得是软件里最难的部分,但确实很酷。

And I think that has created an environment where people are making all of this, right?

而我觉得这就造就了一种环境:人人都在造这些东西。

You give people unlimited tokens.

你给了大家用不完的 token。

Everybody at OpenAI is is very agentic, has great ideas, and so everybody's building everything.

OpenAI 的每个人都非常 agentic,都有很棒的点子,所以人人都在造各种东西。

Whereas I think, you know, you look back at product process that we've all run for a long time, and it's been a little bit opposite, right?

而回头看我们所有人跑了很久的那套产品流程,它其实有点反过来的。

It's been kind of research ideation maybe there was some prototyping but it was you know even when we got past waterfall it was still kind of flavored of like the implementation is expensive and so you what you want to do is you want to derisk all implementation up front through documents through research through prototypes because prototypes and designs are cheaper was kind of the the assumption there uh and that's changed that's like totally changed and right now I'm sure there are 90 different explorations for there's this feature that we desperately need to do that I'm sure there are 90 different uncoordinated teams like implementing and trying, right?

那套流程大致是先做调研、想点子,也许有一些做原型的环节,但即便我们早就走出了瀑布模型,它骨子里还是带着一种味道:实现很贵,所以你想做的是提前通过文档、通过调研、通过原型把所有实现风险都降下来,因为原型和设计更便宜——这大概就是背后的假设。而这一点变了,彻底变了,现在对于某个我们急着要做的功能,我敢肯定有 90 种不同的探索,有 90 个互不协调的团队在各自实现、各自尝试。

Um,

嗯……

so I guess the short answer is like it's it's backwards and it's not that people are doing fundamentally different roles or focusing on different things or that even skill sets have vanished or that roles have just disappeared.

所以简短的回答是:它反过来了,并不是说大家在做根本不同的角色、或关注不同的东西,也不是说技能不再需要了、或者角色就此消失了。

It's that it's backwards, right?

而是整个顺序反过来了。

The taste.

关键在 taste。

Um but it's the curation process.

但关键是那个甄别的过程。

It's like of those 90 attempts like what's good about these?

就是说,在那 90 次尝试里,这些好在哪?

What should we fold into other aspects of this?

哪些该并进这件事的其他部分?

Right?

对吧?

How should we frame this?

我们该怎么去定位、表述它?

Should it be part of this other feature?

它该不该成为另一个功能的一部分?

Right?

对吧?

How many segments should be in the toggle?

那个切换控件里该放几个分段?

Um you know all of those things.

就是所有这些事情。

This episode is brought to you by our season's presenting sponsor, WorkOS.

这期节目由我们本季的冠名赞助商 WorkOS 带来。

What do OpenAI, Anthropic, Cursor, Vercel, Replit, Sierra, Clay, and hundreds of other winning companies all have in common?

OpenAI、Anthropic、Cursor、Vercel、Replit、Sierra、Clay,以及其他数百家成功公司,都有什么共同点?

They are all powered by WorkOS.

它们都由 WorkOS 提供支持。

If you're building a product for the enterprise, you've felt the pain of integrating single sign-on, SCIM, RBAC, audit logs, and other features required by large companies.

如果你在为企业客户做产品,你一定体会过接入单点登录、SCIM、RBAC、审计日志这些大公司要求的功能有多痛苦。

WorkOS turns those deal blockers into drop-in APIs with a modern developer platform built specifically for B2B SaaS.

WorkOS 用一个专为 B2B SaaS 打造的现代开发者平台,把这些拦路的成交障碍变成即插即用的 API。

Literally every startup that I'm an investor in that starts to expand up market ends up working with WorkOS.

我投资的每一家开始向高端市场扩张的初创公司,最后都用上了 WorkOS。

And that's because they are the best.

那是因为它们是最好的。

Whether you are seedstage startup trying to land your first enterprise customer or a unicorn expanding globally, WorkOS is the fastest path to becoming enterprise ready and unblocking growth.

无论你是想拿下第一个企业客户的种子期初创,还是正在全球扩张的独角兽,WorkOS 都是让你最快具备企业级能力、打通增长的那条路。

It's essentially Stripe for enterprise features.

它本质上就是企业级功能里的 Stripe。

Visit workos.com to get started or just hit up their Slack where they have actual engineers waiting to answer your questions.

访问 workos.com 就能开始,或者直接去他们的 Slack,那里有真正的工程师等着回答你的问题。

WorkOS allows you to build faster with delightful APIs, comprehensive docs, and a smooth developer experience.

WorkOS 让你借助好用的 API、完善的文档和顺畅的开发体验,更快地做出产品。

Go to workos.com to make your app enterprise ready today.

现在就去 workos.com,让你的 app 今天就具备企业级能力。

Taste such a such a buzz word.

taste 真是个特别流行的词。

I want to come back to that.

我想待会儿再回到这个话题。

Uh this idea of 90 prototypes.

还有 90 个原型这个说法。

So interesting.

太有意思了。

So just to make sure I understand that.

我只是想确认一下我理解对了。

So there's an idea out there floating around OpenAI.

就是说,OpenAI 内部有一个到处流传的想法。

What people used to do is write docs.

过去大家会做的事是写文档。

Yeah.

对。

Here's what we're going to build.

写清楚我们要造什么。

Here's the feature.

这是功能。

Here's the strategies.

这是策略。

P.

P。

Today what you're describing which makes all the sense is people just create a prototype and what you're saying is people across the company have kind of similar ideas and now instead of a dock they create their little prototype and that leads to kind of 90 different things people can look at and maybe pick here's a direction we want to go down.

而你今天描述的——这完全说得通——是大家直接做一个原型,你的意思是,公司里不同的人有着差不多的想法,现在他们不写文档,而是各自做一个小原型,于是就有了大概 90 个不同的东西摆在那儿供人查看,也许再挑出'这就是我们想走的方向'。

Is that idea?

是这个意思吗?

There's a lot of this um and you know it's not just happening here like you've seen many product leaders say PRDs are dead prototypes are in and I actually don't believe this at all.

现在有很多这种说法,而且不只是这里在说,你看到很多产品负责人都在讲:PRD 已死,原型才是王道,可我其实完全不认同。

I think that one of the interesting things that is happening right now is that because implementation has gotten so cheap across every medium.

我觉得当下正在发生的一件有意思的事是,因为在每一种媒介上,实现都变得如此廉价。

It's very tempting to jump straight to a prototype, especially if you're not an engineer, right?

于是直接跳到做原型就变得很诱人,尤其当你不是工程师的时候。

Especially if you've never been able to write code or never been interested or never had the time.

尤其当你从来没能力写代码、或者从没兴趣、或者从没时间的时候。

It's really tempting to say like PRDs are dead.

确实很容易让人觉得 PRD 已死。

Let me just show you what I mean.

我直接演示给你看就好。

Right?

对吧。

What I've also noticed though is that for engineers, it's really tempting to write a lot of documents, a lot of documents that are not worth reading.

但我也注意到,对工程师来说,写一大堆文档同样很诱人,那些文档很多都不值得读。

Um, this is no shade on people writing documents.

这不是在贬低写文档的人。

It's that if implementation is abundant, then it's really important to pick the right format for the point you're trying to make.

而是说,如果实现是廉价而充裕的,那么为你想表达的观点挑对格式就变得极其重要。

If that point is product clarity around a vague area, then it might actually be a document.

如果这个观点是要把一个模糊领域的产品思路讲清楚,那它可能真的就该是一份文档。

If what you're trying to do is get something in people's hands to try out and to stress test an interaction pattern, it's a prototype.

如果你想做的是把某个东西交到人们手里去试用、去压测一种交互模式,那它就是个原型。

But I think like this is kind of the funny thing now, which is that like it's really important to pick the medium.

但我觉得现在有意思的地方就在这儿——挑对媒介真的很重要。

There's this uh term that a podcast guest shared that I think about when you say this uh which is it's called the primal mark. when a designer or a painter or an artist just creates the first mark on on a painting or a piece of art, that mark is what you start to respond to.

有位播客嘉宾分享过一个词,你这么说的时候我就想到它,叫做 primal mark(第一笔):当一位设计师、画家或艺术家在画作或作品上落下第一笔时,你后续所有的反应都是冲着那一笔来的。

And so everything kind of trickles down from that first mark you make.

于是之后的一切,都从你落下的第一笔层层延伸出来。

And what I'm hearing you saying is sometimes the prototype is the wrong first thing to do because then you're just responding to this prototype versus a different idea versus a bigger idea.

而我听你的意思是,有时候原型是错误的第一步,因为那样你就只是在对这个原型做反应,而不是在回应另一个想法、一个更大的想法。

So I love hearing this.

所以我很喜欢听你讲这个。

So like everyone's just like, "Okay, forget it.

因为现在大家都是那种:好,别管了。

No more writing, no more docs, no more PRDs." You're saying they're actually still useful for specific use cases.

别再写了,别再搞文档了,别再搞 PRD 了。而你是在说,它们对特定场景其实仍然有用。

Yeah.

对。

I I I think too there's this part of the previous world was that the medium implied it had baked in a lot of a lot of signal around where in the process something was right.

我还觉得,过去那个世界有一部分是这样的:媒介本身就内建了大量信号,告诉你某个东西处在流程的哪个阶段。

So if you're seeing something that feels like the app in production that means that it's late in the process that assumptions have been derisked that you know design is looked at this that this is a good business goal right and now those things are sort of divorced right and the reason it was that way is because it was hard to get resources to build the thing until it was properly derised and now that's like just out the window right and so I think it's really important to start saying look we can have prototypes we can have documents Is it are we clear around what this is doing?

所以如果你看到一个东西,感觉像是已经上线的正式 app,那就意味着它处在流程后期、假设已经被排过险、设计已经过目、这是个不错的商业目标,对吧;而现在这些东西彼此脱钩了,之所以以前是那样,是因为在一个东西被妥善排险之前,很难拿到资源去把它做出来,而现在这道门槛直接没了,所以我觉得很重要的是开始说:看,我们可以有原型、可以有文档,但我们对这东西到底在干什么这件事清楚吗?

Right?

对吧。

Because to your point, you do not want to over anchor on this thing that was meant to be an exploration, but now it looks so production ready that like, oh, visually it's ready for prod, but it's not actually the right model of of where the research is going or what users are asking for or what's right for the business, right?

因为就像你说的,你绝不想在一个本来只是探索性质的东西上过度锚定,可它现在看起来太像成品了,让人觉得视觉上都可以上线了,但它其实并不是研究方向、用户诉求或者商业需求的正确模型,对吧?

Not to overdo the taste thing, but it's like it's once again it's like the taste to know like what to work on, how to present that information, like how to achieve the goals, what medium to use is emerging as like the most important thing to do.

不想把 taste 这事讲得太玄,但又一次,就是那种 taste——知道该做什么、怎么呈现这些信息、怎么达成目标、用什么媒介——正在浮现为最重要的一件事。

And that's it.

就是这样。

That's that's in every field.

每个领域都是如此。

What is taste when you talk about good taste?

当你谈 good taste 的时候,taste 到底是什么?

Is it is it what you describe deciding here's the thing we're going to invest in?

是不是你刚才描述的那种判断——决定这就是我们要投入的东西?

Is it also once you have a thing, is this right?

还是说,当你已经有了一个东西之后,它也是指:这样对吗?

Is this the thing to ship? talk about when you think about what is good taste, good judgment, what is what does that concretely?

这是该上线的那个东西吗?讲讲你心里的 good taste、good judgment 是什么,具体到底指什么?

Cuz people hear this word, they're like, "Oh, I have good taste.

因为人们听到这个词就会想:哦,我有 taste。

I know it.

我懂的。

What does it look like in practice?"

它在实践中到底长什么样?

Yeah.

对。

Um, it's funny.

挺有意思的。

There was a tweet I'm too online.

有条推文——我太泡在网上了。

There there was a tweet, I think it was yesterday from the head of product at Linear.

有条推文,我记得是昨天的,来自 Linear 的产品负责人。

I might be getting that wrong.

我可能记错了。

Sorry to anybody who said people overemphasize the aesthetic part of what taste means. and they used um Paul Graham's Craig, but they used him as an example saying Paul Graham clearly has great taste and wears cargo shorts, right?

如果说错了先跟当事人道个歉——大意是人们过度强调了 taste 里审美的那一部分,他们还举了 Paul Graham 当例子,说 Paul Graham 显然 taste 极好,却穿着工装短裤,对吧?

Like, you know, we gota we got to like tease out what taste means a little bit.

就是说,我们得把 taste 的含义稍微拆解一下。

Um and and there's a lot of nuance here.

这里面有很多微妙之处。

I think it's all of the above to to what you mentioned.

我觉得你提到的那些它全都是。

It's it's the like there is an aesthetic part to it.

它确实有审美的那一面。

Um but there's also a a systems thinking part of it.

但它也有系统思维的那一面。

Like how does this fit in the system?

比如这东西如何嵌进整个系统?

There's a where are we going and how like what what theme is this part of?

还有一层是:我们要往哪儿去、这属于什么主题的一部分?

There's how to present this.

还有怎么去呈现它。

A lot of it is wider context and you know obviously there are parts of taste that are like okay this interaction animation doesn't fit in the semantic meaning it's supposed to, right?

其中很大一部分是更宽的上下文,当然,taste 里也确实有一些部分是这样的:好,这个交互动画跟它本应传达的语义并不匹配,对吧?

Like it's too snappy for what it's actually trying to convey.

比如它相对于真正想传达的东西太干脆利落了。

And that's incredibly important and I focus probably too much on that.

这一点极其重要,而我可能在这上面花了太多注意力。

Um, but there's there's like the the like what should what should this be like if we can build anything like what's what's the what's the goal here and how do we how do we get there that I think is like actually the real taste question here.

但还有那种更根本的:如果我们什么都能造出来,这东西究竟应该是什么、目标是什么、我们要怎么抵达那儿——我觉得这才是真正的 taste 问题所在。

When I when I hear things like this I always wonder where will human brains continue to be valuable as AI becomes stronger and better and doing and doing more of of the work and it feels like taste is is a part of it.

每次听到这种说法,我都会想:随着 AI 越来越强、能干的活越来越多,人脑在哪些地方还会继续有价值?感觉 taste(品味/判断力)是其中一部分。

Something I think about along these lines is just AI is still very bad at actual design.

顺着这个思路,我常想的一点是:AI 在真正做设计上还是很差。

Like the output of AI is not great.

AI 出的东西没那么好。

Yeah.

对。

Rarely is it like this is it, they nailed it.

很少会出现那种"就是它了,他们做到位了"的情况。

Y and it's always like, oh, this is Claude design.

而且一眼就能看出来,"哦,这是 Claude 的设计"。

This is Codex design.

"这是 Codex 的设计"。

Why do you think AI and the top frontier models are just not good at design today?

你觉得为什么 AI 和顶尖的 frontier 模型今天就是做不好设计?

Yeah.

嗯。

And do you think they'll get there?

你觉得它们最终能做到吗?

Do you think we'll get to a place of like, holy moly, we're done?

你觉得我们会走到那种"我的天,大功告成了"的地步吗?

Yeah.

对。

I tend to to think that there are some practical reasons why it's lagged and also some harder problems to crack.

我倾向于认为,设计之所以落后,既有一些现实层面的原因,也有一些更难啃的问题。

I'm not in our research or like I'm sure I'll get yelled at for saying this.

我不在我们的研究团队,说这话估计会挨骂。

I think design's a little bit harder to grade um than than software and that you know creating a loop where you can train the model on like what's good design and what's bad design is just a little bit more tedious and ownorous than you know does the code compile does it you know do do what it's supposed to right it because the human aspect of taste is is like part of the feedback mechanism you need I also think that uh the labs historically invest in making their models good at things that accelerates AI research and that in the era the early era of coding models is very clear that the model being able to write correct code would accelerate research right in a way that you can't really make the same case for design.

我觉得设计比软件更难打分,你要搭一个 loop、拿"什么是好设计、什么是坏设计"去训练模型,这件事比"代码能不能编译、有没有实现该实现的功能"要繁琐、麻烦得多,因为 taste 里的人的因素本身就是你需要的反馈机制的一部分;我还觉得,这些实验室历来会投入去让模型擅长那些能加速 AI 研究的事,而在编程模型的早期,"模型能写出正确代码会加速研究"这一点非常清楚,可你很难对设计讲出同样的道理。

Not that getting good at design isn't important, it's that it's not directly in that that flywheel, right?

不是说把设计做好不重要,而是它不直接处在那个飞轮里,对吧?

Um those are practical reasons and I you know those will go away like these models will get pretty good at design.

这些是现实原因,它们会消失,这些模型在设计上会变得相当不错。

There are some kind of murkier things that is is going to be really tough.

但还有一些更模糊的东西,会非常棘手。

Um like I have kind of a short list of them.

我心里大致有个清单。

One is there is an aspect of of culture to what is considered good design in that you remember it was probably what last year where like every new website that came out was just a copy of Linear’s website right like Linear’s website great design great taste like if a model did that I'd be like wow this is incredible leaps here right if I have a model that outputs Linear’s website every time that's not the challenge here right um there's 's an amount of like novelty that is more important in design than it actually is in software engineering.

第一点是,"什么算好设计"里有一层文化的成分,你还记得大概去年吧,每个新出的网站基本都是照抄 Linear 的网站,对吧,Linear 的网站设计很棒、taste 很好,如果一个模型能做出那样的东西,我会觉得"哇,这进步太惊人了",可要是我有个模型每次都吐出 Linear 的网站,那就不是这里真正的挑战了,对吧,设计里有一定的新颖性,它比在软件工程里其实要重要得多。

Like software engineering, you almost you almost want it to overindex unknown patterns, right?

像软件工程,你几乎是希望它去过度依赖已知的模式,对吧?

Whereas design it's like no, there's an element of randomness here and and novelty, right?

而设计则不然,这里有一种随机和新颖的成分,对吧?

There's also the you know to me like I I spent a lot of time writing code or you know supervising code on the early Codex app and even as the models get good at design there's sort of a an abstraction layer that is an interplay between the software design and the code it's being written.

对我来说还有一点,我在早期的 Codex app 上花了大量时间写代码、或者说盯着代码,而即便模型在设计上变强,软件设计和它写出的代码之间还隔着一层抽象,两者相互交织。

Like this thing over here in this corner should share x y and z in the codebase with this thing down here.

比如这个角落里的这个东西,应该和下面这里的那个东西在代码库里共享 x、y、z。

Right?

对吧?

And that's a little bit different than saying the model needs to be a better designer especially on the like you know that's not visual that is visual design but it is significantly deeper.

这跟说"模型得当个更好的设计师"有点不一样,尤其是那部分并不是视觉层面的,它是视觉设计,但要深得多。

It's about the abstractions in that like oh if tomorrow our company did a rebrand.

它关乎抽象,比如,哦,假如明天我们公司做了一次品牌重塑。

The shallow version of this is that we have to you know update 263 components one by one.

浅的版本是,我们得把 263 个组件一个一个更新一遍。

The deep version is like the semantics between these two things that look different like they're both in list that have the like this style that convey this interaction pattern to the user and I think like that is still feeling a little bit out out of reach with the current technology right that abstraction layer.

深的版本是,这两个看上去不同的东西之间的语义,比如它们都在一个列表里、都有这种样式、都向用户传达这种交互模式,而我觉得这一点用当下的技术还是有点够不着,就是那一层抽象。

So I think you know as as we've gone through this process right of you know we started the Codex app in uh November and we weren't using it full-time now we use it for everything that's been a journey but now it's like the things that we actually do while using it are different things.

所以我觉得,我们一路走下来是这样的,我们 11 月启动了 Codex app,当时并没有全职用它,现在什么事都用它,这是一段历程,但如今我们用它时真正在做的事,已经是另一些事了。

So what was the question?

所以刚才的问题是什么来着?

I know that was that was an amazing answer.

我知道,那是个特别棒的回答。

Uh, speaking of design and being creative, the Codex app when it came out, it's like such a new thing.

说到设计和创造力,Codex app 刚出来的时候,是个特别新的东西。

Yeah.

对。

That nobody has had seen before.

以前没人见过的那种。

It's like not a terminal thing.

它不是个 terminal 的东西。

It's not an ID thing.

也不是个 IDE 的东西。

It's like this chat thing that codes and you could see code.

它像是这种会写代码、又能让你看到代码的聊天式的东西。

To your point, it feels like it'd be a hard for AI to be like here's a whole new paradigm for how to code.

按你刚才说的,感觉让 AI 拿出"这是一套全新的编程范式"会很难。

And that feels like where human brains continue to be valuable for now is like creativity almost and coming up with something new versus like a patterns of things that have been done before.

而这感觉正是眼下人脑还继续有价值的地方,几乎就是创造力,是拿出全新的东西,而不是那些以前已经做过的模式。

Yeah.

对。

I mean, I totally agree.

我完全同意。

Let's give it up for the human brain

为人脑鼓个掌吧。

for now.

暂时是这样。

As we were getting ready for this, you said that you were listening to the episode of Jenny, who is the head of design for Claude Code and co-work and such, and she had this whole kind of thesis that the design process is dead.

我们准备这期节目的时候,你说你在听 Jenny 那一期,她是 Claude Code、co-work 之类项目的 head of design,她有一整套观点,认为设计流程已经死了。

There's no time for design.

没时间做设计了。

Things are moving too fast.

一切都变得太快。

Uh, just build now and design is kind of steering things as things move along.

现在就是先造出来,设计只是在推进的过程中做些把舵的活。

You're implying you have kind of a different perspective on the design process.

你话里的意思是,你对设计流程有一套不太一样的看法。

We probably agree on on a lot of this, Jenny and I.

这件事上我和 Jenny 大概有很多共识。

I wasn't a fan of the design like the design process proper.

我本来就不是那套正儿八经的设计流程的拥趸。

I agree with with her take that it is it is dead and I genuinely was not a fan of this process before AI like I I

我同意她的判断——它确实已经死了,而且早在 AI 出现之前我就 genuinely 不喜欢这套流程,我

can you describe the process real quick just from people think about the design.

你能先快速描述一下这个流程吗,就是大家想到设计时会想到的那套。

So I mean I when I read a startup a number of years ago, we we would you know do design hiring and there was this sort of snarky article that came out about like the case study factory and it was it was like midsurp era stuff, right?

我意思是,几年前我在一家创业公司的时候,我们要做设计岗的招聘,当时出了一篇挺刻薄的文章,讲的是所谓的“case study 工厂”,那大概是设计工具井喷的那个年代的东西吧?

And it was that designers are being taught about this process and valuing that above all else, above all outcomes even, right?

文章说的是,设计师被灌输这么一套流程,并把它奉为高于一切、甚至高于一切结果的东西。

And if something went through this process that two things were true. one, it would be good and the process would guarantee quality and guarantee impact and also that if something the thing was good if it went through that process even if you don't like it and nobody uses it.

而只要某个东西走过了这套流程,就有两件事成立:第一,它会是好的,流程能保证质量、保证影响力;第二,只要它走过了这套流程,它就是好的,哪怕你并不喜欢它、哪怕根本没人用。

It's like the process was you know of of user research and the divergence and the convergence.

就是那套用户调研、发散、收敛的流程。

It's the right framework.

它是对的框架。

It was always a little academic, but I think I think this is really exposing some areas where it falls down, especially because of the speed of implementation.

它一直都有点学院派,但我觉得,这真的暴露出了它站不住脚的一些地方,尤其是因为实现的速度变了。

And like once again, like that process is sort of predicated on the assumption that implementation is expensive and that you can really only afford to build once.

再说一次,那套流程某种程度上是建立在一个假设之上的:实现是昂贵的,你真的只能负担得起造一次。

And so you need to fully like exhaustively go through the problem space and the solution space before implementing, right?

所以你必须先彻底、穷尽地把问题空间和解空间走一遍,然后才去实现。

And that you and and then like kind of we saw with like you know Figma and Origami and all of these tools that you can fast forward some of the insights by pulling interactive prototypes earlier into the process, right? that you can, you know, simulate production and like, you know, there ended up being sort of a meme about executives just being like, well, can we just do a prototype and then like expect it to, you know, work.

然后我们看到,像 Figma、Origami 这些工具,你可以把交互原型更早地拉进流程里,从而提前拿到一部分洞察;你可以模拟出接近生产环境的效果,最后甚至衍生出一个梗:高管们就会说,那我们能不能直接做个原型,然后指望它就这么上线跑起来。

But, but this thing was real, right?

但这件事是真实存在的。

That this became part of the the design process proper, right?

它成了正儿八经设计流程的一部分。

We pulled prototyping into that.

我们把做原型拉进了流程里。

The problem now is that you can pull all of the implementation into that.

现在的问题是,你可以把整个实现环节都拉进来。

And there's a mismatch between I think a lot of assumptions again like you see this fully polished prototype that looks like it's ready to go out the door and enough people at a company see that and they're like can we can we release this now but the appropriate like we're actually in that early design process stage and nobody's just saying that right like this is this is where we are with like a bunch of like multiplayer exploration right you know 90 people will have this idea it'll look really polished but it's like no this is actually that that's the design process now, right?

这里存在一种错配,又是很多假设的问题:你看到这个完全打磨好的原型,看起来已经可以直接出门上线了,公司里够多的人看到它,就会说“我们现在能不能发了”,但恰当的说法应该是——我们其实还在设计流程的早期阶段,只是没人这么讲罢了;这就是我们现在所处的状态,一堆人在做 multiplayer 式的探索,90 个人会冒出这个想法,它看起来会非常精致,但其实——不,这才是如今的设计流程。

Tying the design process to mediums, media, uh like that's the scary part.

把设计流程和媒介、载体绑死,这才是可怕的地方。

It's that designers have more tools now to do this process with, right?

现在设计师有了更多工具去跑这套流程。

You can put stuff into the current product and you can A/B test it or just, you know, use that as a prototype.

你可以把东西直接放进现有产品里做 A/B 测试,或者干脆拿它当原型用。

Um many companies right now have this idea of like a baby version of the product like baby cursor.

现在很多公司都有这么个想法,就是搞一个产品的婴儿版,比如 baby cursor。

You've seen this on Twitter like we have baby codecs, right? a dramatically simplified codebase that approximates all of the interactions of the production app and therefore is a lot quicker to vibe code over, right?

你在 Twitter 上见过的,我们有 baby codex——一个被大幅简化的代码库,近似复刻了生产环境 app 的所有交互,因此在上面做 vibe coding 要快得多。

Because you can be like, well, what if the sidebar worked like this or what if a pain came in and had like a group chat here?

因为你可以这么试:要是侧边栏这样运作会怎样,或者要是弹进来一个面板、这里带个群聊会怎样?

What if XYZ, right?

要是这样那样呢?

That's like a huge tool that's part of the design process.

这就是设计流程里一个巨大的工具。

So to say the design process is dead, I feel like it's both true and false, right?

所以说设计流程死了,我觉得这既对也不对。

It's that if you are if you are tied to the tools in the ex like the spec the exact like day-to-day specifics of the process then yeah it's dead like you're not going to have a good time but to throw the process out completely or throw like the overlay of the process the like hey we're at this point in the process like that is still more important than ever it's

如果你被那套工具、被那份 spec、被流程里精确到日常操作的细节绑死了,那没错,它死了,你会很难受;但要把整套流程彻底扔掉,或者把流程的那层框架也扔掉——就是那种“嘿,我们现在处在流程的这个节点”的意识——那玩意儿反而比以往任何时候都更重要,它

it's really interesting because you have a background in every function if people look at your LinkedIn it's like engineer designer product manager founder uh now You oversee the desktop app and I think design is not under your perfume.

这真的很有意思,因为你什么职能的背景都有,大家去看你的 LinkedIn,上面写着 engineer、designer、product manager、founder,现在你负责桌面 app,而且我觉得设计不归你管。

Is that right?

对吗?

Is there like a separate design team or are they under your

是有一个独立的设计团队,还是他们归在你

depends on the week?

取决于是哪一周?

Okay, good.

好,不错。

We work very closely together.

我们合作得非常紧密。

Like we believe in all sitting together being embedded like I reporting lane.

我们信奉大家坐在一起、深度嵌入彼此,比如汇报线上——

I don't

我不

they sh the shift weekly.

他们每周都在变。

What does the design process look like on on the Codex?

在 Codex 上,设计流程是什么样的?

Yeah, there's been a lot written about ro collapse, existential role collapse.

嗯,关于 role collapse、关于生存性的角色融合,已经有很多文章写过了。

There are no roles anymore.

角色不复存在了。

We haven't seen that.

我们没看到这种情况。

Um we've we have seen more role collapse in the Codex org um than I think other parts of the company and other parts of the economy.

但我们在 Codex 团队里确实看到了比公司其他部门、比经济里其他领域更多的角色融合。

I think part of this is that this was a technical product for engineers and so our designers speak engineer right our product managers speak technical language and write code.

我觉得部分原因在于,这是一个面向工程师的技术型产品,所以我们的设计师会讲工程师的语言,我们的产品经理会讲技术语言、会写代码。

Alexander has a master's degree in computer science, which is I do not have a master's degree in computer science.

Alexander 有计算机科学的硕士学位,而我可没有计算机科学的硕士学位。

Like, so we've seen a lot of role collapse and and I think that, you know, one of one of the ways that we describe how the groups work together is that there's significantly more overlap in the roles than there used to be.

所以我们看到了大量角色融合,而我觉得,我们描述这些小组如何协作的方式之一,就是各角色之间的重叠比过去多了很多。

And everybody's sort of defined less by the fence and the boundaries of where design stops and engineering starts, but more the average of where they're working, right?

每个人越来越不是由那道栅栏、由“设计到哪儿为止、工程从哪儿开始”的边界来定义,而更多是由他们工作重心的平均位置来定义。

So like, you know, if you average up all of the things that somebody on our design team does, there's plenty of codew writing things.

所以说,如果你把我们设计团队里某个人做的所有事情平均一下,里面有不少是写代码的活。

There's plenty of things that are product work, but on average, like their dots over here, right?

也有不少是产品的活,但平均下来,他们的点大概落在这一带。

If you draw you draw it out on a diagram, and this sort of speaks to the process, too.

如果你把它画在一张图上——这其实也说到了流程本身。

Um, especially because the entirety of the Codex app has been informed by the dog fooding loop.

尤其是因为整个 Codex app 都是被 dogfooding 这个 loop 塑造出来的。

There is a desire among all of us to try to do as much as possible in the app even when it's not the best tool so that it can become the best tool.

我们所有人都有一种渴望:尽可能多地在 app 里完成工作,哪怕它此刻还不是最趁手的工具,好让它最终成为最趁手的工具。

And so a lot of design we all work on by using the app and say okay what's broken about this?

所以很多设计工作,我们都是靠亲自用这个 app 来做,然后问:这地方哪里不对劲?

This is a whole thing we do which is that we we often don't improve our process so that we can make the product better to do it which is a deeply like uncomfortable place to be in but you know week to week it's changing.

这是我们一直在做的一整件事:我们常常不去优化自己的流程,好让产品本身在这件事上做得更好——这是一个让人极其不适的处境,但你知道,它每周都在变。

I love this point so much that uh like what are you?

我太喜欢这个点了——就是,你到底是什么?

It's your role is the average of what you spend your time on.

你的角色,就是你把时间花在什么上的那个平均值。

If most of your work is PM work, then okay, you're a PM for now.

如果你大部分工作是 PM 的活,那好,你眼下就是个 PM。

If it's engineering, you're an engineer for now.

如果是做工程,那你现在就是工程师。

I feel like was OpenAI the first company to uh call people member of technical staff.

我感觉,OpenAI 是不是第一家管员工叫 member of technical staff 的公司?

No, I I believe that this this might have started with Xerox.

不,我记得这个叫法可能是从 Xerox 开始的。

First company I interned at, it was a company called Up There did the same thing.

我实习的第一家公司,叫 Up There,也是这么叫的。

This has been around, but it's it's it's much more common now. you know, it is kind of a tradition in research focused companies,

这个头衔早就有了,只是现在更常见,算是研究导向型公司的一种传统。

right?

对吧?

Okay, got it.

好,明白了。

So, it emerged from research, but I feel like it's such a I don't know sign of where things might be headed.

所以它是从研究领域里冒出来的,但我觉得这特别像是一个信号,预示着事情可能的走向。

This idea of uh we're just going to call everyone me member of technical staff.

就是那种想法——我们干脆把所有人都叫 member of technical staff。

Your function isn't set.

你的职能不是固定的。

You're not like in this bucket of the PM or theorg of the design or do you feel like that's where we all head long term?

你不会被塞进 PM 那个格子,或者设计那个部门,你觉得我们长期都会走到这一步吗?

Do you feel like functions will continue to exist?

你觉得职能划分还会继续存在吗?

Like there's still the PM skill set and the NG skill set and the design skill set.

比如仍然有 PM 的技能组、工程的技能组、设计的技能组。

And people are I'm a designer or do you think this is like like people call it builder?

人们会说"我是个设计师",还是说你觉得会像有些人讲的那样,大家都叫 builder?

Do you feel like there's this collapse coming where everyone's everything and that's just the future or do you think we're going to continue to be mostly divided up in?

你觉得会不会有一场角色融合正在到来,每个人都无所不做,这就是未来;还是说你觉得我们仍然会大体上按职能分工?

There are there are some things that I'm afraid of.

有些事情是我担心的。

And I think that, you know, some some companies like to be very extreme about getting onto the bandwagon of whatever people say is going to happen.

我觉得,有些公司特别喜欢走极端,别人说什么将要发生,他们就一头扎进去追那股风。

And I think part of the danger in eliminating the concept of roles is that it can dangerously eliminate the idea that things are specialties with knowable best practices, right?

我觉得取消角色这个概念,危险的地方之一在于,它可能会危险地一并抹掉这样一种认知:很多事情是有门道的专业,是有一套可习得的最佳实践的,对吧?

I I've heard a lot of companies be like, we're getting rid of the product role, which I think is, by the way, a terrible idea.

我听到不少公司说,我们要把产品这个岗位砍掉——顺便说一句,我觉得这是个糟糕透顶的主意。

Um, and everybody's just going to be like a builder.

然后所有人就都变成所谓的 builder。

And then what happens is they don't like this whole discipline of product that's been built up and has like real best practices, real things that have been tried and failed and like real processes like that just gets abandoned because people are like, "Oh, I wrote some code, right?" Like that's not a great place to be in.

接下来的结果是,产品这一整套已经积累起来、有着真正最佳实践、有着无数试错、有着成熟流程的学科,就这么被抛弃了,因为大家觉得"哦,我写了几行代码嘛",对吧?这可不是什么好状态。

I think that the boundary of like so and so like this isn't your lane.

至于"这不归你管、这不是你的地盘"这种边界,

I I welcome that part going away, but there's a balance here where it's like not everyone can work on everything for one, both in terms of breath and depth, right?

这一部分我倒乐见它消失,但这里有个平衡:一来,不是每个人都能什么都干,无论从广度还是深度上讲,对吧?

Like this is why managers are not going to go away.

这也正是管理者不会消失的原因。

Um, not everybody can work on everything and also like every discipline has a skill component to it, which I think a lot of engineers are guilty of not recognizing that like well engineering is a skill to it. it's writing code and like other roles are just people vibing.

不是每个人都能什么都做;而且每个学科都有它的技能门槛,我觉得很多工程师的毛病就是不承认这一点——好像只有工程才是门手艺、才是写代码,别的角色不过是些人在那儿凭感觉瞎晃。

It's like no that's not how it works, right?

这就好比说,不,事情不是这样运作的,对吧?

Um like yes, you can use Excel you but you cannot work on the finance team, right?

打个比方,是的,你会用 Excel,但你不能因此就进财务团队干活,对吧?

That is like that kind of stuff, right?

就是这类事,对吧?

Yeah.

嗯。

I think there's also just like do you want to be doing this work?

我觉得还有一层是,你到底想不想做这份活儿?

Do I want to be

我想不想去

I think more of it that actually now is like it's easier to switch roles.

我现在更倾向于这么看:换角色变得更容易了。

It's easier to learn the best practices.

学习最佳实践变得更容易了。

It's easier to not tie your effectiveness in a role with the ability to use the exact tool, right?

把你在一个角色里的战斗力,不再和会不会用某个特定工具绑死,也变得更容易了。

It's more of like can you get yourself into this mindset, learn which things work and which don't and then like focus on it, right?

现在更关键的是,你能不能让自己进入这种心态,搞清楚哪些做法有用、哪些没用,然后专注地去做,对吧?

Like I I spent so long feeling like I should not be a software engineer because I didn't care about like assembly language or memorizing TypeScript syntax, you know?

就像我曾经很长一段时间都觉得自己不该当软件工程师,因为我根本不在乎汇编语言,也不愿意去背 TypeScript 的语法。

And it's like there have always been parts of these roles that are that sort of gatekeeping that are like, well, no, this like being good at this role is being good at this tool.

这些角色里一直都存在着某种设卡把关的成分,就是那种"不行,想在这个角色里做得好,就得先精通这个工具"的东西。

And I think that's what's kind of starting to erode.

而我觉得,正在开始瓦解的就是这一部分。

I just don't I I think people take this they hyperbolize all of this.

我只是觉得,人们把这一切都夸大其词了。

What does your team look like on on on the Codex team?

在 Codex 团队里,你的团队是什么样子的?

How many engineers, designers, PMS?

有多少工程师、设计师、PM?

What's kind of like the makeup of the team right now?

现在团队的构成大概是怎样的?

Every time people ask me like how many people are on the Codex team, do you remember my answer to this?

每次有人问我 Codex 团队有多少人,你还记得我的回答吗?

Yeah, I I'm like it's somewhere between 10 and a few thousand.

对,我会说,大概在 10 人到几千人之间。

I mean, it's like a fake answer, but it's it's real in that we do see this as the culmination of

这算是个不正经的回答,但它其实是真的,因为我们把这个产品看作整个公司工作的集大成者——

what everybody works on here.

所有人在这里做的一切,最终都汇入了它。

Like everything that goes into model research, everything that goes into how you know models are good at CUA and browser use, everything about how you know model personality, all of the product work around you know front-end infrastructure, all of the user like all of it is this product.

模型研究的每一分投入,模型在 CUA 和浏览器操作上做得好的每一分投入,模型人格的方方面面,围绕前端基础设施的所有产品工作,所有用户层面的——这一切,都是这个产品。

At the same time we are not accepting PRs daily from thousands and thousands of people on whatever they want.

但与此同时,我们并不是每天都在接收成千上万人随心所欲提交的 PR。

Um so we got a team uh double digits of engineers probably half that on the design side. um you know few product people although you know product tier is kind of more of his own defense play and I think one thing that is very common around among everybody on the Codex side or or on the desktop side is agency and taste right a lot of former founders or people who were at larger companies doing founder shaped things um a lot of people with with immense taste at OpenAI we let teams get very large so we haven't that hey there's no management but like the teams are quite large right it's mostly IC's um and I think that's good

所以我们有一支团队,工程师是两位数,设计这边大概是它的一半,产品的人不多——不过产品这一层更像是各自为战的打法。我觉得 Codex 这边、或者说桌面端这边所有人身上一个很共同的东西,就是 agency 和 taste,对吧?很多是前创始人,或者曾在大公司做过创始人那种活儿的人,OpenAI 里有很多 taste 极强的人。我们放手让团队变得很大,所以我们并没有说完全没有管理,而是说团队规模相当大,对吧?大部分都是 IC,我觉得这挺好的。

you use this term zone defense for product work and I think that's really interesting it kind of maps to the design kind of shift also just like you're there to kind of manage and coordinate talk a little bit more about what that looks like what does zone defense look like for a product person

你用了区域联防这个词来形容产品工作,我觉得这特别有意思,它也和设计那种转变对上了——你在那儿更多是做管理和协调。多讲讲这具体是什么样子吧,对一个产品人来说,区域联防是什么样的?

yeah and I I have had a lot of conversations with with Alexander about this this analogy which is that like if two product people are working too closely that's often not a good signal And that like you kind of want like as a product or you sort of want to do this like forced directed activity where you're like where are the gaps especially in this new world where curation and like you know steering and alignment is a lot of things where you're like there's a ton of chaos happening on people throwing ideas all over the place right the whole like top down you know year-long planning thing not going to work and so now it's like we need the taste makers to guide things from incept ception to what the product should be.

嗯,我跟 Alexander 就这个比喻聊过很多次。意思是,如果两个产品人配合得太紧密,那往往不是个好信号。作为产品人,你其实想做的是一种像力导向布局那样的活动,去找哪里有空档——尤其在这个新世界里,甄别、掌舵和对齐是很大一部分工作,因为到处是混乱,大家把想法四处乱抛,对吧?那种自上而下、以年为单位的规划完全行不通了。所以现在是,我们需要 taste maker 从最初的构想一路引导,把产品该有的样子带出来。

And that means you basically want company coverage.

这也就意味着,你基本上想要覆盖整个公司。

And so you spread out and you say, "All right, who's like who's best at what?

于是你们分散开,说:好,谁最擅长什么?

Let's create some space between us so that we got full coverage, right?" And that's kind of how it goes.

咱们彼此之间留出些空间,这样就能全场覆盖,对吧?大致就是这么运作的。

And then you fill in the gaps and you're like, "Look, like we want to hire engineers who are product minded.

然后你去把空档补上,你会说:听着,我们想招那种有产品思维的工程师。

Like we don't we don't want it to be that, you know, we've got a bunch of people writing a bunch of code that needs like full team reviewing it for like product coherence, right?

我们不希望是这样:一堆人写着一堆代码,还得靠整个团队去审它的产品连贯性,对吧?

Like we want everyone to have these skills, but I think like what people go deep on has to change, right?

我们希望每个人都具备这些技能,但我觉得,人们往哪里钻深必须得变,对吧?

This is definitely a thread I've been noticing over and over with talking to folks like you is the the most valuable person right now, one of the most valuable is someone that could take an idea from idea to done with the taste to know this is great.

跟你这样的人聊,有一条线索我一次又一次地注意到:当下最有价值的人——之一——是那种能把一个想法从点子做到完成、并且有 taste 去判断这东西很棒的人。

Just like shephering throughout this obsession with making it awesome like this kind of high agency, high taste person.

他会一路呵护它,痴迷于把它打磨得出色,就是这种高 agency、高 taste 的人。

Exactly. as you described, is that is that kind of the way you think about here's who we're hiring, here's who's going to do really well in this new world?

正是如此。就像你描述的那样,这是不是你们思考该招什么人、谁会在这个新世界里做得特别好的方式?

Yeah, I think that that's that's the core piece right now.

对,我觉得这就是眼下最核心的一块。

And it it also speaks to how I sort of see IC versus management, which is that it's not that management is going away.

它也呼应了我对 IC 和管理关系的看法:并不是管理要消失了。

It's not that everyone's an IC, but like everyone's kind of both now, right?

也不是说人人都成了 IC,而是现在每个人某种程度上两者都是,对吧?

If you're an IC, you're not typing code out character by character, right? like you are managing something you're managing agents you're managing you know like you're managing work that is happening right that comes together to do a certain thing if you're a manager of teams you're doing the same thing just at a different like different granularity right I generally look for like obviously command over the discipline but then the taste to say like hey you're going to have unlimited tokens and I don't like we can't just be doing slop like you need to be able to determine what's signal, what's noise, like in a world of just infinite content.

如果你是 IC,你不再是一个字符一个字符地敲代码,对吧?你是在管理某样东西,你在管理一堆 agent,你在管理正在发生、并最终汇聚成某个成果的工作。如果你管的是团队,你做的是同一件事,只是粒度不同,对吧?我一般会找那种显然对本专业有掌控力的人,同时还要有 taste,能说:嘿,你会有用不完的 token,而我们不能只是在产出 slop,你得能判断什么是信号、什么是噪音——在一个内容无穷无尽的世界里。

Mentioned planning uh at the pace things are moving.

你提到了在这种变化速度下做规划。

It's become very hard to plan road maps.

现在制定路线图变得非常困难。

Yeah,

是啊。

I imagine especially in our world.

我想在我们这个领域尤其如此。

Uh

呃,

people are very frustrated with me all the time on this.

大家一直因为这件事对我很不爽。

Yes.

对。

Because things are just constantly shipping, things are changing, right?

因为东西一直在往外发,一直在变,对吧?

How do you plan on on on your team?

你在你的团队里是怎么做规划的?

What's kind of like how far ahead are you thinking and what does a plan look like?

你大概会往前想多远,一份计划长什么样?

Is it like a spreadsheet?

是一张表格吗?

Is it an MD file?

还是一个 MD 文件?

What what's kind of the output of a plan?

一份计划最后产出的是什么?

Yeah, I I don't think we do anything revolutionary on that.

嗯,我觉得我们在这方面没做什么革命性的东西。

We're not clever about planning.

我们在规划上并不高明。

I think like the basic gist is the shorter term something is, the more detail it needs.

我觉得基本的意思是:越是短期的事,越需要细节。

And then it's not that we don't plan for 9 months out, it's that that just has to stay very hazy because any amount of precision that you add to a 9-month plan right now is false precision.

并不是说我们不为 9 个月以后做规划,而是那部分必须保持非常模糊,因为你现在给一份 9 个月的计划添加的任何精确度,都是虚假的精确度。

And like you're just going to waste time.

你只会白白浪费时间。

Like you can say stuff, right?

你当然可以说点什么,对吧?

But like nothing that we planned.

但我们计划的东西没一样会成真。

I think research is different.

我觉得研究那边不一样。

So I'm not speaking for research here but like on the applied side when we do product like anything that you could have planned in November may have been true for December but like isn't what happened right so it's hard like it is really hard to do planning we generally need to know like what do we think models are able to do on what timeline and my last company I kind of saw this shift where we were starting to use the models to drive features and the product process fell down.

所以我这里不代表研究那边说话,但在应用这一侧、我们做产品的时候,任何你在 11 月能规划出来的东西,放到 12 月可能还成立,但根本不是实际发生的样子,所以真的很难、做规划真的非常难,我们通常需要知道:我们认为模型在什么时间点能做到什么,而在我上一家公司,我算是亲眼看到了这种转变——我们开始用模型来驱动功能,原来那套产品流程就垮掉了。

It basically had to be like, let's list out all of the things that we think we are interested in doing for the next year or two.

基本上必须变成:我们把接下来一两年里我们感兴趣想做的所有事情都列出来。

Let's prototype all of them, decide which things are ready now, and then just let the others sit and bake, and then every time there's like a new leap in models, let's try that thing again with it swapped out. because like the whole premise of whether features were good or not were based on whether they were smart enough, not the shape of them.

把它们全部做成原型,判断哪些现在就能上,然后让其余的先搁着慢慢酝酿,每次模型出现新的飞跃,就把那一块换掉、再拿新模型试一遍。因为功能好不好的整个前提,取决于模型是否足够聪明,而不是功能本身的形态。

So this is a great story about the Codex app.

关于 Codex app,有个特别好的故事。

I like I am very confident that the Codex app that we released in February, if that had been ready in November, it would have absolutely failed in the market and that that the only difference was the models between November and February, right?

我非常有把握:我们二月发布的这个 Codex app,如果放到十一月就做好、拿出来,那它在市场上一定会惨败,而唯一的区别就是十一月到二月之间的模型,对吧?

And I I think like there's a lot to that that this product with the exact same shape I think would have like it's it's outcomes were totally different depending on just a few months of timing.

我觉得这里面很有讲究——同样形态的这个产品,仅仅因为时间差了几个月,结果就完全不同。

This episode is brought to you by Mercury.

本期节目由 Mercury 赞助。

Radically different banking loved by over 300,000 entrepreneurs and now with command.

一种彻底不一样的银行服务,已获得超过 300,000 名创业者的喜爱,现在还带上了 Command。

I've been a customer of Mercury's for over 6 years.

我做 Mercury 的用户已经超过 6 年了。

I have never once thought about leaving.

我从没有一次动过要离开的念头。

Mercury is basically what happens when banking is built by product people, not by bankers.

Mercury 本质上就是——当银行由做产品的人、而不是银行家来造时,会是什么样。

They make it so easy, dare I say fun, to send invoices, move money around, set up virtual cards for folks on my team.

他们把开发票、转账、给我团队的人设置虚拟卡这些事做得如此简单,我敢说甚至有点好玩。

Does your bank have an API, a terminal native CLI or an AI ready MCP server?

你的银行有 API 吗?有原生终端的 CLI 吗?有为 AI 准备好的 MCP server 吗?

I don't think so.

我看是没有。

And just recently, they launched Command, a conversational interface built directly into Mercury, which acts as your financial operator.

而且他们最近刚推出了 Command,一个直接内建在 Mercury 里的对话式界面,充当你的财务操盘手。

I've been using Command to transfer money around to figure out what categories I've been spending the most money in, analyze my cash flows, and just today I used it to find out how much I've made from a specific sponsor over the past year.

我一直用 Command 来转账、搞清楚自己在哪些类别花钱最多、分析我的现金流,就在今天,我还用它查了过去一年从某个特定赞助商那里赚了多少钱。

I just ask, "How much have I made from X over the past year?" 10 seconds later, I have an answer.

我只要问一句「过去一年我从 X 那里赚了多少?」10 秒后,答案就出来了。

It is so freaking cool.

这也太酷了。

Visit mercury.com to learn more and apply online in minutes.

访问 mercury.com 了解更多,几分钟就能在线申请。

Mercury is a fintech company, not an FDIC-insured bank. banking services provided through Choice Financial Group and column NA members FDIC.

Mercury 是一家金融科技公司,不是 FDIC 承保的银行。银行服务由 Choice Financial Group 和 Column NA 提供,两者均为 FDIC 成员。

This is definitely a thread on this podcast is build things that are not yet working then will work when the model gets better.

这档播客里绝对贯穿着一条主线:先造那些还跑不通、但等模型变强后就能跑通的东西。

And there's this kind of other thread of ambition.

另外还有一条关于野心的主线。

Be more ambitious with the things you take on.

对你要接下的事,要更有野心。

So is this just like a way you approach things just like let's just build a bunch of things that may not work yet.

所以这就是你做事的方式吗——先造一堆现在还不一定能成的东西。

We'll just have them around and wait for a model to catch up.

先摆在那儿,等着模型追上来。

Is that kind of the approach?

大概是这个路子吗?

Yeah, I think we have a lot of that.

对,我觉得这样的东西我们有很多。

I think sometimes the challenge is like you have to be very clear again about what stage of the design process that's in.

我觉得有时候难点在于,你得再一次非常清楚地界定:它处在设计流程的哪个阶段。

People still have this muscle memory of like, oh, I wrote the code for this thing, therefore we should put it out there.

人们还是有这种肌肉记忆:哦,我都把这东西的代码写出来了,那我们就该把它发出去。

It's like, no, no, no, that means you have an artifact now that we can test against for into future models, right?

而实际是,不不不,那只意味着你现在有了一个 artifact,可以拿它去对着未来的模型做测试,对吧?

Um, this happened with the inapp browser in the app that we have, right?

这事就发生在我们 app 里那个内置浏览器上,对吧?

Like we had a a kind of a working version.

我们当时有个大致能用的版本。

I mean, go back to Atlas, we had agent working inside of Atlas and, you know, that was pretty cool.

往回说到 Atlas,我们在 Atlas 里让 agent 跑起来过,那挺酷的。

We had operator before that in ChatGPT, right?

在那之前我们在 ChatGPT 里有过 operator,对吧?

That didn't work out.

那没成。

Very cool idea.

点子很酷。

Like there's some thread that you can draw between operator atlas Codex ChatGPT that's like fundamentally the same feature, but the re-releasing of it with different intelligence totally changes the outcome here.

你能在 operator、Atlas、Codex、ChatGPT 之间牵出某条线,它们本质上是同一个功能,但换上不同的智能水平重新发布一次,结果就彻底不一样了。

And so I push people not to be stubborn about like no this isn't working so it's a bad feature.

所以我会推着大家别太固执,别觉得「这东西不行,所以它是个烂功能」。

Like no, it might not be ready yet.

而是,不,它可能只是还没到时候。

Um there's also this aspect of especially in research there's always a desire to be the most ambitious and to say okay but at the limit the model can just do this and it just doesn't work on the product side like if you if you go back to the original Codex release basically what it was is it said it was codex web and it wasn't good for interacting with it was like you give the model a task and it's going to go off do the task come back to you with finished like doesn't sound that radical.

还有一个层面,尤其在研究里,总有一种冲动想做到最有野心,想说「行,极限情况下模型就能干这个」,可在产品这边它就是行不通——你回头看最早的 Codex 发布,基本上就是所谓的 Codex web,它并不适合去互动,那种形态是:你给模型一个任务,它就跑开去做,做完带着成品回来找你,听着好像没那么颠覆。

The problem is like it didn't do the task that well.

问题是它任务做得没那么好。

Like it wrote code.

它是写了代码。

It was it was it was good but it was like that form factor was too early and then the Claude Code comes out totally local like not hooked up to the cloud.

写得还行,但那个形态太早了,然后 Claude Code 出来了,完全本地、根本没接云端。

Um doesn't pretend to be as it's not as AGI-pilled right it's like going to ask you questions it's going to sit there you can't just delegate your life to it.

它也不假装自己那么行,它没那么 AGI-pilled,对吧,它会反过来问你问题,会停在那儿等你,你没法把自己的活儿一股脑全甩给它。

That worked way better right because that's the point that the models were at there right?

那反而好用多了,对吧,因为那才是当时模型所处的位置,对吧?

So we were like we were too AGI-pilled for the moment.

所以说,对那个时刻而言,我们太 AGI-pilled 了。

And I think like I I think about that lesson a lot on this stuff.

我觉得这个教训,我在这类事情上想了很多。

It used to be that you know Bailey and market told you all these things about the shape of the product about the communication of the product and now it's like no you might need to release this thing six different times before it works and that might like the shape might not change at all.

过去是,你知道,过去靠品牌和市场信号就能告诉你关于产品形态、产品沟通的一切,而现在是,不,你可能得把这东西发上六次它才跑得通,而它的形态可能压根不用变。

There's like it's so interesting to hear about all the variables you have to think about building product.

听你讲做产品要考虑的这么多变量,真的太有意思了。

Now there's the timeline for the models and the research and how smart it gets. there's like people's ability to even understand this is how you could build software in the cloud and this is the future like get people prepared for this new future and then just uh what you can build as a team and I love that Codex example cuz comes back to this idea of ambition and I want to hear if there's anything there for you of just this threat of just be more ambitious because these models can do so much more than you even imagine and sometimes it's too ambitious for the market and they're not ready for it but you think about that at all just like pushing your team to be more ambitious because it's so much easier to just do things that may be felt crazy hard in a fast.

现在有模型和研究的时间线、它会变得多聪明;还有人们能不能理解「原来软件可以在云端这么造、这就是未来」——得让大家为这个新未来做好准备;然后就是你作为一个团队能造出什么。我特别喜欢那个 Codex 的例子,因为它又回到了野心这个想法上。我想听听,对你来说这里有没有什么东西,就是「要更有野心」这条线——因为这些模型能做的远比你想象得多,有时候对市场来说又太有野心、市场还没准备好——你会想这个吗,就是推着你的团队更有野心,因为如今要做那些以前会觉得难到发疯的事,已经容易太多、也快太多了。

Yeah, this is a core challenge.

对,这是个核心挑战。

Once once there's a product that exists or a feature that exists, it's really easy for people to find paper cuts and

一旦有了一个已存在的产品或功能,人们就非常容易挑出小毛病、去

optimize

做优化,

and they should and people on Twitter like to remind us of this and I thank them for that.

他们也应该这么做,Twitter 上的人爱提醒我们这些,我也为此感谢他们。

Like people should be focused on the features that exist and making them more reliable and better.

人们确实应该盯着已有的功能,把它们做得更可靠、更好。

But this, you know, this is why we also have a culture of bottoms up exploration here because sometimes in the same way the Codex app came and disrupted ChatGPT in some way, right?

但这也正是我们这里同时保有一种自下而上探索文化的原因,因为有时候,就像 Codex app 出现、在某种程度上颠覆了 ChatGPT 一样,对吧?

This thing will get disrupted by a future effort and that's that's part of the design is that like you can't always as one team be good at both the disruptive piece and the like maintaining a product and its quality piece.

这东西日后也会被某个新的努力所颠覆,而这本身就是设计的一部分——作为同一个团队,你没法既擅长做那块颠覆性的事,又擅长维护一个产品和它的质量。

Some point to design a process that allows for both.

总得设计出一套能同时容纳这两者的流程。

kind of zooming out a little bit.

稍微往后拉远一点看。

If you think about the progression that we've been on of uh AI impacting how we build product, it's like insane how far we've come from, as you said, we used to write all our code by hand, like artisally created human code to AI writing 100% of our code to you actually put it this way that now like coding is steering the AI and like when you think about what percentage of my code is written by AI, it's almost like how many times did I have to steer it in the right direction is the AI version of coding.

如果你想想我们一路走过的这条 AI 影响我们如何做产品的进程,那真是疯狂——正如你说的,我们从曾经全靠手写代码、像手工艺一样人工敲出来的代码,走到 AI 写 100% 的代码,再到你实际上是这么说的:如今写代码就是在给 AI 掌舵,而当你去想「我的代码里有多大比例是 AI 写的」时,它几乎变成了「我得把它往正确方向掰多少次」——这就是编程的 AI 版本。

And now that there's like agents and loops and all these things, what's kind of the latest frontier from what you've seen of how people are building is is it loops?

如今又有了 agent、loop 这一大堆东西,从你看到的情况来说,人们做产品的最新前沿大概是什么?是 loop 吗?

Is there something else of just like the most AI AI forward teams?

还是有别的什么——就是那些最 AI 先行的团队?

Here's how they operate now that people may not be aware of.

他们现在是这么运作的,而这可能是大家还没意识到的。

Yeah, I mean loops are so last week, man.

是啊，loop 早就过时了，老兄。

I mean, we we talked about this.

这个我们聊过。

Um, you know, one of the big questions is always, well, how much of the product is AI written?

一个老生常谈的大问题就是：产品里有多少代码是 AI 写的？

And it's always hard to answer that question cuz if you're using the goalposts from last year, it's like, well, 100% of our product right now is AI written code.

而这问题总是很难回答，因为如果你还拿去年的标准线来衡量，那答案就是：我们现在产品里 100% 的代码都是 AI 写的。

So the question is more like, well, okay, fine.

所以真正的问题更像是——好吧。

Is is the code written supervised versus unsupervised, right?

这些代码是在有监督下写的，还是无监督下写的？

And that's like a totally different thing.

这就完全是另一回事了。

I welcome the moving of goalposts because that means we're making product progress.

我乐见标准线被不断移动，因为这说明我们的产品在进步。

There have been a lot of explorations here around like autonomous autonomously developed software. um a lot of like harness engineering stuff, a lot of different explorations.

这方面已经有很多探索，围绕着软件的自主开发，很多 harness 工程的东西，各种不同方向的尝试。

I'm like, okay, well, what if you came in overnight and did garbage collection of the codebase to clean it up, right?

我就在想：要是它能在一夜之间对代码库做一次垃圾回收、把它清理干净呢？

One thing that I think all models suffer with right now is just they they usually increase complexity.

我觉得当下所有模型都有一个通病，就是它们通常只会增加复杂度。

If research is listening at any company, please make the models better at deleting code.

如果哪家公司做研究的人正在听：拜托，让模型更擅长删代码。

Um but you know, that becomes a problem right now when you try to put development completely on autopilot.

但当你想让开发彻底进入自动驾驶状态时，这就成了个问题。

Um, and it's both on the the human side and the codebase side.

而且这在人这一侧和代码库那一侧都存在。

So like feature requests, right?

比如功能需求。

You know, how do you teach a model which features to build, which ones to ignore, which ones to kind of like group together and reframe a little bit?

你怎么教模型去判断哪些功能该做、哪些该忽略、哪些该归并到一起、稍微重新组织一下？

How do you teach a model how to build the right abstractions, right?

你怎么教模型去搭建正确的抽象？

Like all of this is getting better.

这一切都在变好。

Um, I don't think we're at a place yet where we're like, we're just got to set up a loop that's like improve the app, you know, and listens to Twitter and listens to Slack and listens to email and like we're not there yet, but we are we are we're trying to make it happen.

我不觉得我们已经到了这一步：只要架起一个 loop，让它自己去改进 app，去听 Twitter、听 Slack、听 email——我们还没到那一步，但我们确实在努力让它成真。

Do you think we'll get there?

你觉得我们能走到那一步吗？

Do you think we'll get to a place where it's just like grow like win

你觉得我们会走到这样一个阶段吗——就是直接一句「去增长、去赢」

slashgoal make money like make me a billion dollars?

或者「去赚钱，帮我赚十亿美元」？

Win win the market.

赢下这个市场。

I don't know, man.

我不知道，老兄。

Like I

我……

I am not in the business of saying

我不是那种会去说

never or always or whatever.

「绝不」或「一定」之类话的人。

Yeah.

嗯。

How are you using AI in your work as product leader, engine leader?

作为产品负责人、工程负责人,你在工作里是怎么用 AI 的?

What are some ways that you use it that maybe people may not be aware they can use the app for?

有哪些用法是大家可能还没意识到这个 app 能做到的?

Yeah, I I think I have the best job in in the world right now.

是这样,我觉得我现在的工作大概是全世界最棒的。

Um, but one of the things that makes it very fun is that when we were developing the original Codex app, the goal for me personally was to make it the thing that I wrote the code with, right?

而其中一件特别好玩的事是:当初我们做第一版 Codex app 时,我个人的目标就是让它成为我写代码时用的那个工具。

I was like, I need to make this so good at development that I can build the code app with this.

我当时想的是,我得让它在开发上强到我能用它来构建 Codex app 本身。

And the Codex app at that time was a development tool, right?

而那时候的 Codex app 是一个开发工具。

And we did that like super quick dog fooding loop because you got your personal dog fooding loop where you're like, oh, like I can't do this thing.

我们跑的是那种超快的 dogfooding loop,因为你有一个属于自己的 dogfooding loop:你会发现,哦,这件事我做不了。

I should fix that. so that I can do the thing.

我该把它修好,这样我才能做这件事。

Now I can do the thing.

现在我能做这件事了。

Now I can do more things, right?

现在我能做更多事了。

Um you know we released that and then the next challenge was hey people are starting to do some different shaped things with this right and now I you know need to grow this and so I need to hire a few people and help.

然后我们把它发布出去,接下来的挑战是:大家开始拿它做一些形态不一样的事情,那么我就得把它做大,所以我得招几个人来帮忙。

So then like my role changed at the same time that the role of the app needed to change.

于是我的角色发生变化的同时,这个 app 的角色也需要变化。

So I'm like okay I need to do more product discovery here.

所以我想,好,我得在这儿做更多的产品探索。

I need to figure out the right loops for seeing what everybody's working on and steering things that are off track.

我得摸索出合适的 loop,去看清每个人在做什么,并把偏离轨道的东西拉回来。

And so all of a sudden that's what I started using the Codex app for, right?

于是一下子,这就变成了我开始用 Codex app 去做的事。

I did still write code.

我确实还在写代码。

Like I've I've tried to align my own usage of it with the problem that we're trying to solve, right?

我一直试着让自己对它的使用,和我们正要解决的问题对齐。

And now I'm like, I need to build a spreadsheet that models this out.

而现在我会想,我得做一个电子表格来把这件事建模出来。

I need to kind of do a, you know, internal deep research on all of the efforts that have gone into this area of research for the next version of this.

我得做一次内部的 deep research,把之前投入到这块研究领域的所有工作梳理一遍,好为下一个版本做准备。

There's a release or a series of releases in Mayish that introduced the inapp browser computer use and artifact creation to the Codex app.

在五月前后有一次发布,或者说一系列发布,给 Codex app 引入了应用内浏览器的 computer use 和 artifact 生成能力。

That was I think our Codex's for almost everyone release and everybody knows the term vibe coding.

那我记得就是我们那次"让几乎所有人都能用上 Codex"的发布,而 vibe coding 这个词现在人人都知道了。

I think that was like our first five coordinated release where like I had a a you know Notion doc somewhere with everything that needed to happen and I was like automating like going out to gather updates from pull requests from Slack channels and like updating the status tracker and like now this is pretty common place.

我觉得那是我们第一次做多人协同的发布:我在某个地方有一份 Notion 文档,列着所有需要发生的事,然后我把一些事自动化了,比如去 Slack 频道里从 pull request 里收集进展、更新状态追踪表,而现在这些已经相当常见了。

Um but at the time I felt like I was at the bleeding edge of like how to manage a product release.

但在当时,我感觉自己站在"如何管理一次产品发布"的最前沿。

In short, like the way that I use the Codex app is basically like what what has my job grown into and how do I make this thing able to do everything I need to do?

简单说,我用 Codex app 的方式,基本上就是:我的工作成长成了什么样,我就怎么让这个东西能替我完成我需要做的一切。

Um, I will get up in the morning, I will see the daily brief that I have from like everything from the 3,000 Slack channels that I'm in, like which things need my attention.

我早上起来,会看到我的每日简报,它把我所在的那 3,000 个 Slack 频道里的一切都归拢过来,告诉我哪些事需要我关注。

I can kind of message back and be like, "All right, give me five questions and I'll answer them and I can, you know, do that."

我可以回一句消息,比如说"好,给我五个问题,我来回答",然后我就能这么做。

How do you set that up? what's like the workflow to for somebody to set that up because that sounds amazing.

你是怎么设置的?对别人来说,搭这么一套的流程是怎样的?因为这听起来太棒了。

Again, I I think we're still in the like discovery phase on on a lot of this.

还是那句话,我觉得这里面很多东西我们仍处在探索阶段。

And and so right now it's like I'm making an automation that says or scheduled task that says like I go through my Slack channels.

所以现在的做法是,我做一个自动化,或者说定时任务,让它去过一遍我的 Slack 频道。

These are the things that I care about and think are most important, you know.

这些是我在意的、我认为最重要的东西。

So I'm still kind of defining that like these are things to watch out for different categories like here's some context and you know I'll I'll you know that'll get set up as a automated task and then first few times it runs it might need some steering and luckily with this app you know I don't have to find out how to edit the instructions I can just be like hey next time this runs like can you please worry about this instead or can you deemphasize this workstream or hey this thing happened and it didn't come up in a brief like can you make sure that stuff in the shape so I can kind coach it along the way.

所以我还在慢慢定义:这些是要留意的点,分成不同类别,这里给一些上下文,然后它会被设置成一个自动化任务;头几次运行时可能需要一些引导,而幸运的是,用这个 app 我不用去搞清楚怎么编辑那些指令,我可以直接说,嘿,下次运行时能不能改成关心这个,或者把这条工作线的权重调低,再或者,嘿,发生了这么件事而它没出现在简报里,能不能确保这类东西被纳入进来——这样我就能一路把它带着调教。

It'll update the way that it it notifies me, stuff like that.

它会更新通知我的方式,诸如此类。

Amazing.

太棒了。

I think in the future though, like this is this has been a core problem with the the chatbot shape, right?

不过我觉得在未来,这一直是聊天机器人这种形态的一个核心问题。

Is that I know how to set this up.

那就是:我知道怎么搭这套东西。

I have time to set this up because for me it's product discovery to set it up.

我有时间去搭,因为对我来说搭建本身就是产品探索。

But if you're not working at OpenAI, not developing this, like you don't want to have to figure out all this stuff.

但如果你不在 OpenAI 工作、不是在开发这东西,你根本不会想去搞清楚这一堆事。

Like we we we need to like figure out that that shape of things.

我们得把那种形态摸索出来。

Yeah.

嗯。

What I'm hearing is I don't think people realize that uh your app can act a lot like open claw.

我听下来的感觉是,我觉得大家没意识到你们的 app 可以很像 open claw 那样工作。

You was was people were so excited about like you just talk to it, set up this thing, check on this thing for me every day and then tell me what's going on like like it's starting to become a part of all these all the products.

大家特别兴奋的一点就是:你只要跟它说话,把这件事设好,让它每天替我盯着这件事,然后告诉我发生了什么——它开始变成所有这些产品的一部分了。

Uh which is amazing.

这太棒了。

So the way somebody would set this up is they just talk within the app and say I want to set up an automation to do this.

所以别人要搭这套,做法就是在 app 里直接对它说话:我想设一个自动化来做这件事。

Look at my Slack and these are the things I I want to do.

看我的 Slack,这些是我想做的事。

Yeah.

对。

Great.

很好。

Yeah.

对。

And the app will say you know

然后 app 会说,

Yeah.

对。

It'll set it up for you.

它会替你设置好。

If it doesn't have a Slack connector, it'll say, "Can I add the Slack connector?" Yes or no?

如果它还没有 Slack connector,它会问:"我可以加上 Slack connector 吗?"是或否?

You can hit yes to that.

你可以点"是"。

Like the least that we can do is make it so that if you don't know how to do something in the app, they can just ask it, right?

我们至少能做到的一点是:如果你不知道在 app 里怎么做某件事,你直接问它就行。

Yeah.

对。

I don't think that's enough, but I think that's the least we can do.

我不觉得这就够了,但我觉得这是我们至少能做到的。

Yeah.

对。

A good example is I built this little app that filters spammy email forbox.

一个不错的例子是,我做了一个小 app,用来过滤收件箱里的垃圾邮件。

So every email that comes in and I built this in Codex.

每一封进来的邮件——这个我是用 Codex 做的。

Uh every email that comes in, it looks at it and decides is this uns unsolicited kind of cold email stuff that I don't want to look at and labels it and puts it somewhere else.

每一封进来的邮件,它都会看一眼,判断这是不是那种我不想看的、未经许可的陌生推销邮件,给它打上标签,挪到别处去。

And to set that up, one of the steps was you have to go into like the Google cloud console and set up all these Pub/Sub API things and triggers.

而要把这套设好,其中一步是你得进到 Google Cloud Console 里,把这一堆 Pub/Sub API 之类的东西和触发器全设好。

I don't know if you've ever used that interface.

我不知道你有没有用过那个界面。

It's like so annoying and slow, right?

它又烦又慢。

So I was like, wait, what if I ask you to do it?

所以我就想,等等,要是我让你来做呢?

And I was like, okay, cool.

我就想,好,酷。

Do this for me.

替我做这件事。

And it just you describe computer use.

然后它就——这正是 computer use。

I've never actually seen this happen on my computer before.

我以前从没真正在自己电脑上见过这种事。

It just takes over my computer and starts going there.

它直接接管我的电脑,开始往那儿操作。

It's like I don't care if you don't have a connector, man.

那感觉就像:哥们,我才不管你有没有 connector。

I'll just start clicking.

我直接开点。

Yeah.

对。

And it figures it out.

然后它就把事情搞定了。

It's crazy just to like watch it doing its thing.

光是看着它自己捣鼓,就觉得离谱。

Designing the decision boundary between connectors.

设计 connector 之间的决策边界。

When to use the inapp browser versus your Chrome extension that's connected versus computer is interesting and all done through just like feeling it out.

什么时候用应用内浏览器、什么时候用你连好的 Chrome 扩展、什么时候用 computer——这一点很有意思,而且全靠自己一点点摸出来。

I saw a great Twitter thread the other day where they describe all these three and what you use it for.

我前几天看到一个很棒的 Twitter thread,里面把这三种都讲了一遍,以及各自用来干什么。

So, this person described it really well.

这个人讲得真的很好。

These personal workflows are really interesting because some of them really click like some of you know people are trying all sorts of stuff.

这些个人化的工作流特别有意思,因为其中一些真的很对味,大家都在各种各样地尝试。

Everybody's making these personal systems.

每个人都在搭这些属于自己的系统。

You ask everybody here what they do and everything's going to be different and then like certain themes arise and we're like you know what that should be a first a first class experience on the app. like we should take this thing that everybody seems to be setting up and just make that work.

你问这儿的每个人他们都怎么用,答案各不相同,然后某些共同的主题浮现出来,我们就会想,你知道吗,这应该成为 app 里的一等体验,我们应该把这个大家似乎都在搭的东西直接做成开箱即用。

And I think memory is sort of in the shape where we've had a lot of people and a lot of people at other companies too are like well I set up an Obsidian base or a Notion area and I tell it how to basically build my mind palace and how to put it's like I don't know if everyone like you shouldn't have to do that like there should be a memory feature that does that for you right that's pretty generic.

我觉得 memory 大概就是这种形态:我们碰到很多人,别家公司也有很多人,都会说,我搭了一个 Obsidian 库,或者一块 Notion 区域,我告诉它怎么去搭建我的记忆宫殿、怎么往里放东西——我就觉得,我不确定是不是每个人都该这么干,你本不该非得这么做,应该有一个 memory 功能替你把这事做了,这挺通用的。

So that's you know that's one but then there are other things like they're your process of your job and there's something that like yeah you should set that up but I think this is we're constantly waiting through like what's working for individual people like what should what should enter the product versus stay like no that's just how you do your job right what should become a primitive what shouldn't um

所以那是其中一类,但还有别的东西,比如那是你这份工作的流程,有些东西确实该你自己去搭,但我觉得我们一直在不停地筛:对个人有效的东西里,哪些该进产品,哪些就该留着——不,那只是你干活的方式而已,哪些该变成一个原语,哪些不该。

yeah this is the the taste and judgment you spoke of earlier you know citing these things I want to talk about this browser use piece a little bit because I think people don't realize how powerful this is and what it could be used for.

对,这就是你前面说的那种 taste 和判断力,把这些东西挑出来。我想稍微聊聊浏览器使用这块,因为我觉得大家没意识到它有多强大、能拿来做什么。

Reminds me, I don't know if you watched uh when Dan Shipper was on the podcast, he had this from every he had this prediction that we're going to start using Codex to run our SaaS apps inside.

这让我想起,不知道你有没有看 Dan Shipper 上播客那一期,他有一个来自 Every 的预测:我们会开始用 Codex 在里面运行我们的 SaaS app。

So, instead of going in Chrome,

所以不用再开 Chrome,

I know he slacks me about this every day asking for stuff.

我知道他每天都在 Slack 上跟我念叨这事,问我要各种东西。

Do you feel like this is where things go where we're just working within the Codex app using notion and linear and Salesforce inside with your agent kind of helping you along or do you think that's a kind of a different direction?

你觉得事情会往这个方向走吗——我们就在 Codex app 里面干活,在里面用 Notion、Linear、Salesforce,让你的 agent 一路帮着你;还是说你觉得那是另一个方向?

It's been Yeah, it's been really interesting because obviously we've had a few attempts at browser shaped activity, right?

这一直——是的,这一直特别有意思,因为显然我们在浏览器形态的活动上已经试过好几次了。

And operator and tragic de agent mode and atlas and now we have the inapp browser inside of the desktop app.

有 operator,有(某个)agent 模式,有 Atlas,现在我们在桌面 app 里有了应用内浏览器。

We also have the the ability to install a Chrome extension where the app connects to Chrome.

我们还有安装 Chrome 扩展的能力,让 app 连到 Chrome。

Like we've had a lot of shapes at this and I think we've we've learned a lot of different things.

我们在这件事上试过很多种形态,我觉得我们学到了很多不同的东西。

There's there's a lot at play.

这里面有很多东西在起作用。

There's a lot of really boring things at play.

有很多非常无聊的东西在起作用。

Like you know, we originally launched the app.

比如,我们最初发布这个 app 时。

It's an Electron app.

它是一个 Electron app。

The things that you can do with inapp browsers and there it's it's like kind of janky.

你能用应用内浏览器做的那些事,其实有点糙。

So we had the the inapp browser was for development.

所以我们当时的应用内浏览器是给开发用的。

It was for testing your front end on development.

它是用来在开发时测你的前端的。

And we were like it's not really for anything else, guys. like it's it's a developer tool, right?

我们当时的态度是,伙计们,它真的不是拿来干别的的,它是个开发者工具。

And then we we switched over to our owl owl stack which had powered the Atlas browser and so now you know multitab and we've got enterprise security so that you can actually log into all your websites if you're you know so uh we've been iterating on this.

然后我们切换到了我们自己的 owl 栈,就是驱动 Atlas 浏览器的那套,所以现在有了多标签页,我们还有了企业级安全性,这样你就真能登录你所有的网站了——所以我们一直在迭代这块。

I think the tough thing has always been what should the shape of this browser be like is this something that is only for the agent right that like you've got Chrome you open Chrome you you know do your thing in Chrome if you ask the desktop app it opens up this browser that it can control really quickly doesn't have the latency of Playwright whatever but that you know or are we trying to say this app is for everything and like we want you to use this as a browser and those have a lot of trade-offs it's not super you know welltraveled path, right?

我觉得一直以来最棘手的问题是:这个浏览器到底该是什么形态,它是只给 agent 用的吗——就是说你有 Chrome,你打开 Chrome,在 Chrome 里做你的事;而如果你去问桌面 app,它会打开这个它能很快控制的浏览器,没有 Playwright 那种延迟之类的——还是说我们想讲的是这个 app 什么都能干,我们希望你把它当浏览器用?这两条路各有很多取舍,它不是一条走得很熟的路。

Like most browsers are browsers at the top level that got browser tabs.

毕竟大多数浏览器,顶层就是个浏览器,有浏览器标签页。

Um, this creates a lot of really boring but tedious problems like keyboard shortcuts, right?

这会带来一大堆非常无聊但又麻烦的问题,比如键盘快捷键。

Are we trying to do key mapping to VS Code or to Chrome or to our own thing or to linear or like, you know, we want to have some sort of muscle memory that carries over but got all these things that have shape like subshapes of different products out on the market.

我们是要把键位映射到 VS Code,还是 Chrome,还是我们自己的一套,还是 Linear?我们希望有某种能延续下来的肌肉记忆,但市面上又有各种各样成形的产品、各种子形态。

What do we do?

我们该怎么办?

And this just highlights how extra challenging this app is where you have to allow it to work for somebody that's never built anything

而这恰恰凸显了这个 app 有多加倍地难:你得让它既能服务一个从没构建过任何东西的人——

from like the more basic user to like power like Peter

——从比较基础的用户,一直到像 Peter 那样的高阶用户,

Open Claw trying to code with it.

用 open claw 试着拿它来写代码。

I I'm not convinced I'm going to get Peter to use the app.

我可没把握能让 Peter 用上这个 app。

I think he might be the last terminal terminal hold out.

我觉得他可能是最后一个死守终端的人。

Okay.

好吧。

But I'm going to I'm going to keep trying.

但我会继续、继续努力。

Okay.

好。

Let me uh let me zoom out for a moment and talk about kind of the big picture of where you're taking all this.

我想往后退一步,聊聊你把这一切带向何方的大图景。

What's the what's kind of the vision for for Codex?

Codex 的愿景是什么?

Where does this go?

它会走向哪里?

What's it going to look like?

它会长成什么样子?

You know, I don't know, a year or two, 10 years.

我不知道,一两年后、十年后。

We had Codex as a CLI, right?

我们先有了作为 CLI 的 Codex,对吧?

And then we decided to build this app.

然后我们决定做这个 app。

And, you know, we were we were a little uncertain about the app and um but had a lot of conviction in what it could be as to start a developer tool, right?

我们对这个 app 其实有点没底,但对它作为一款起步的开发者工具能成为什么,我们有很强的信念。

That

就是这样,

and it wasn't going to be an IDE.

它不会是一个 IDE。

It was going to be this right-sized surface where it was like sort of a chatbot, but it was more than that and you could see the code, but we weren't going to let you edit the code.

它会是一个尺寸恰到好处的界面,有点像聊天机器人,但又不止于此;你能看到代码,但我们不会让你编辑代码。

There's a really interesting thing that happened at OpenAI in January in February, and it was before we actually released the Codex app, which is that we'd started to dog food the Codex app.

1 月到 2 月,OpenAI 内部发生了一件很有意思的事,那是在我们真正发布 Codex app 之前,就是我们开始 dogfooding 这个 Codex app。

And what we were finding like we were converging on some pretty clear internal PMF on engineering right and research workflows.

我们发现,在工程和研究工作流上,我们已经收敛出了相当清晰的内部产品市场契合(PMF)。

They were thrilled.

他们欣喜若狂。

They were loving it.

他们爱不释手。

We were like all right we just got to get the quality bar up before we release it to the world.

我们想,好,那我们只要在向全世界发布之前把质量门槛提上去就行。

We're convinced that this will be a thing.

我们确信这会成气候。

But then at the company we spun up a few other workflows to say hey like this Codex effort is onto something with these coding agents and we have people from marketing from comms from finance from legal from basically every discipline who are using this Codex app even though it is actively hostile to these people right it is like trying to show them code it's trying to ask for approval to run RG on the you know it's like it's doing all of these things that are actively not the right product surface for them.

但接着在公司内部,我们又拉起了几个别的工作流,想说:嘿,这个 Codex 的方向,靠这些 coding agent 摸到了点门道;结果来自市场、传播、财务、法务、基本上每个职能的人都在用这个 Codex app,哪怕它对这些人是充满敌意的——它一直想给他们看代码,想申请权限去跑 rg,它一直在做各种对他们来说根本不对路的产品动作。

So why don't we take our other surfaces and add Codex to them?

那我们为什么不把我们别的界面拿来,给它们也加上 Codex 呢?

Like let's add it to the ChatGPT desktop app.

比如把它加进 ChatGPT 桌面 app。

Let's add it to the atlas browser, right?

把它加进 Atlas 浏览器,对吧?

And let's essentially take the lessons of Codex and make it more general for a general knowledge work tool, right?

本质上就是把 Codex 的经验拿过来,把它做得更通用,变成一个通用的知识工作工具,对吧?

And those efforts went for a little bit and the the most annoying problem happened which is nobody would leave the Codex app for the apps that were allegedly for these other personas.

那些尝试推进了一小段,结果最烦人的问题出现了:没人愿意离开 Codex app,去用那些号称是给这些其他人群准备的 app。

And I think the lesson in all of this was just that like the whole developer tool versus general knowledge work tool like there's a lot of nuance here that isn't just one or the other.

我觉得这一切给出的教训就是:开发者工具和通用知识工作工具之间,有很多微妙之处,并不是非此即彼。

And I think we really we believe really strongly in this and that there are certainly in the same way that we talk about the average of your role is like what your role is now.

我觉得我们真的非常笃信这一点——正如我们说你角色的平均值就是你现在的角色,

This is true on the product side too.

在产品这一侧同样成立。

People who are doing Excel work don't want to see git repository information.

做 Excel 工作的人不想看到 git 仓库信息。

We know that.

这个我们清楚。

But we also know we can tell a lot from what they're doing about what kind of work they do and we can start simple grow the product complex as we feel is needed.

但我们也清楚,从他们在做的事里,我们能读出很多关于他们做哪类工作的信息;我们可以从简单起步,再按需要把产品长复杂。

Right?

对吧?

Doesn't mean we don't have modes, right?

这不代表我们不搞模式,对吧?

You might want some modes for organizing your stuff and to sort of be like legible about the the ways that you enter the experience, right?

你可能会想要一些模式来组织你的东西,让你进入这个体验的方式变得清晰易读,对吧?

But we really believe strongly that what we've built here is the right shape to take take on like really deep vertically focused things, right?

但我们真的非常笃信,我们在这里造出来的东西,是承接那些真正纵深、垂直聚焦的场景的正确形态。

We work deeply with our finance team with our team working on science team working on legal, right?

我们和财务团队深度协作,和做科学的团队、做法务的团队深度协作,对吧?

And and we say like if we can build the right extensibility primitives in the right general model, then you can do anything with this, right?

我们会说:如果我们能在正确的通用模型里,搭出正确的可扩展性原语,那你用它什么都能做,对吧?

And then our challenge is well how do you you know how do you generalize it?

然后我们的挑战就是:那你要怎么把它通用化?

But this is kind of going back to like the best desktop app that we can build like what does that look like?

但这又绕回到:我们能造出的最好的桌面 app,它长什么样?

Um and so you know was Codex the developer tool ChatGPT like where does this go?

所以你看,是 Codex 这个开发者工具,还是 ChatGPT,这一切走向何方?

This is how we think about it.

这就是我们的思考方式。

It is so interesting the point you made that the Codex app was so like you did such a good job getting people to be aware it existed and so good to use and fun to use that everyone's starting to use that versus the ChatGPT app.

你提到的那个点特别有意思:Codex app 你们做得太成功了——让大家知道它存在、好用又用着开心,以至于所有人都开始用它而不是 ChatGPT app。

So clearly the direction is combine them so that you're not creating this confusion which I know is you know things people have been talking about this idea of com bringing them together.

所以方向显然就是把它们合起来,免得制造这种混乱;我知道把它们合到一起这个想法,大家已经在讨论了。

Somebody called it a super app and wish they hadn't said that cuz now I have to hear about the super app all day every day.

有人管它叫超级应用(super app),真希望他们没这么说,因为现在我每天从早到晚都得听到这个超级应用。

We'll get past it.

我们会熬过去的。

Great.

很好。

Okay.

好。

But is that is that kind of the not let's not call it a super app but the idea is like one place people go to do all the things.

但那是不是——我们先别叫它超级应用——但那个想法是不是就是:一个地方,人们去那里把所有事都办了。

Is that the general idea or TBD?

这是大致的想法,还是待定?

Yeah.

对。

I think what we see here is that it's a great home base.

我觉得我们在这里看到的是,它是一个很棒的 home base。

It's a great place to keep track of all of the things that you have to do across different surfaces and some of those things you do all of it in the app.

它是一个很棒的地方,帮你追踪你在不同界面上要做的所有事情;其中有些事你完全在 app 里做完。

Some of those things the app opens other apps to to do right the app can connect to Excel so that you know yes it has a spreadsheet editor inside the app.

另一些事,app 会打开别的 app 去完成——app 可以连到 Excel,是的,它 app 内部自带一个电子表格编辑器。

Is that good enough for people doing financial modeling at OpenAI for raising billions of dollars?

对那些在 OpenAI 为了融资几十亿美元而做财务建模的人来说,那个够用吗?

Like probably not.

大概不够。

Um, and so the app talks directly to the addin in Microsoft Excel on your desktop.

所以这个 app 会直接跟你桌面上 Microsoft Excel 里的插件对话。

When it's done, you can close Excel, right?

做完之后,你可以把 Excel 关掉,对吧?

And so it's not just about, hey, we're drawing a rectangle on the screen and everything needs to happen in that rectangle.

所以这不只是说:嘿,我们在屏幕上画了一个矩形,一切都得在这个矩形里发生。

It's this thing should be a home for you where you start work, you end work, you automate work, and it uses whatever you need to do.

而是这个东西应该成为你的一个家,你在这里开始工作、结束工作、把工作自动化,它会调用你需要的一切。

Right?

对吧?

There's a there's a great story about we had some videos that we shot in this room for the original launch of the Codex app and our in-house DX videographer Brent then was tasked with editing all these videos, right?

有个很棒的故事:我们为 Codex app 的最初发布,在这个房间里拍了一些视频,然后我们内部的 DX 视频师 Brent 被派去剪辑这些视频,对吧?

And he edited all the videos with Codex which was one of the early like whoa what are people doing with this thing, right?

他用 Codex 剪完了所有视频,这是早期那种"哇,人们竟然拿这东西干这个"的案例之一,对吧?

And the process for why he decided to start using Codex was really interesting.

而他决定开始用 Codex 的过程特别有意思。

He started just because he was curious if Codex could edit videos.

他一开始纯粹是好奇 Codex 能不能剪视频。

And so Codex is not a video editor per se, right?

Codex 本身并不是一个视频剪辑器,对吧?

It doesn't have any of that UI in it, but um it was able to understand that he used Premiere Pro.

它里面没有任何那种 UI,但它能理解到他用的是 Premiere Pro。

It could do some edits by editing the files that were backing what was on screen in Premiere Pro, but it couldn't do everything.

它可以通过编辑那些支撑 Premiere Pro 屏幕内容的文件来做一些剪辑,但它没法做全部。

So naturally, what Codex then did was built itself an extension that could be installed into Premiere Pro that it could then talk to and say, "Hey, Premiere Pro extension, can you please change this marker inside of the Premiere Pro app." That was pretty nuts when we saw that happening.

于是很自然地,Codex 接下来做的,是给自己造了一个能装进 Premiere Pro 的扩展,然后它就能跟这个扩展对话说:"嘿,Premiere Pro 扩展,能麻烦你把 Premiere Pro app 里的这个标记改一下吗。"我们看到这一幕发生的时候,真是相当疯狂。

Um, it's a great model, right? there are these like specialty tools that specialize in things and so we're trying to do two things at once with Codex and with now with ChatGPT one is how can we seamlessly interact with these tools that you're already using and say like we don't we don't need to build a better video editor for you right but like Codex and ChatGPT can use that video editor right it can interact can hand stuff off to it right so how can we do that and that's often through um connectors or computer use or even extensions in this case right And then there is, you know, Dan Shipper’s thing, which is, hey, I have these web apps that you can click around and use, but I want to be able to open these in codecs and have Codex do extra stuff with it, right?

这是个很棒的模型,对吧?有这些专精于某些事的专业工具,所以我们想用 Codex、现在还有 ChatGPT,同时做两件事:一是我们怎么能无缝地跟你已经在用的这些工具交互,说我们不需要给你造一个更好的视频剪辑器,但 Codex 和 ChatGPT 可以用那个视频剪辑器,它可以跟它交互、可以把活儿交接给它,那我们怎么做到这一点——这往往是通过 connector、computer use,甚至像这个例子里的扩展来实现的,对吧?然后还有 Dan Shipper 提的那个想法,就是:嘿,我有这些可以点来点去、直接用的 web app,但我想能在 Codex 里打开它们,让 Codex 拿它们做些额外的事,对吧?

And so those are kind of like two models that are almost inverse of each other that we're doing a lot with both at the same time.

所以这算是两种几乎互为反向的模型,我们同时在这两边都下了很大功夫。

This premiere story is interesting to me because it's another example of just be more ambitious with these AI jobs.

那个 Premiere 的故事我挺感兴趣,因为它又是一个例子——面对这些 AI 任务,就该更大胆一点。

Like you may may not know, maybe they could do this thing.

就是你可能不知道,也许它们能做成这件事。

It's almost just like go try go try out see if it figures it out.

几乎就是去试试、去试一下,看它能不能自己搞定。

I'm going to take us to a recurring corner on the podcast that I call fail corner.

我要带我们进入播客里一个固定环节,我管它叫失败角。

And so the question for you is people see people like you just like killing it, just growing.

所以问题是:大家看到像你这样的人一路势如破竹、一路在成长。

Everything's winning.

什么都在赢。

Codex is doing so great.

Codex 做得这么好。

This crazy career, everything's up and to the right.

这条疯狂的职业路,一切都在往右上方走。

People may not see the times that things didn't work out and things that you launched that were failures.

大家可能看不到那些不顺的时候,看不到你上线过的那些失败的东西。

And so these stories are really important for people to hear that it's not all just a win all the time.

所以这些故事对大家来说真的很重要,让人知道并不是一直都在赢。

What's a story of a time you failed in your career that taught you something really important?

能不能讲一个你职业生涯中失败、却教会了你某些真正重要东西的经历?

It's funny to hear that description played back at me and like this is perhaps the first time I've I've not felt like I was failing.

听你把这段描述回放给我,挺好笑的,因为这大概是我第一次没觉得自己在失败。

I mean, I was a startup founder for a long time.

我当了很长时间的创业公司创始人。

I ended up selling the company for parts essentially, right?

最后基本上是把公司拆开卖掉了。

And it was just it was years.

那折腾了好些年。

It was a slog.

是一场苦役。

It was heavily regulated spaces.

都是监管很重的领域。

The whole thing felt like a constant failure.

整件事感觉就是接连不断的失败。

Um I went to this other startup and we were trying to do some AI tools and in this also like pretty locked down regulated industry and that felt like just time after time of trying things and it not working.

我又去了另一家创业公司,我们想做一些 AI 工具,那也是个相当封闭、受监管的行业,那感觉就是一次又一次地尝试,又一次次不成。

So to me it's been like I've failed actually quite a lot and you know sometimes it's just a point in time where things line up like skill set passion point in the market line up we know with this project to bring what we've learned with the Codex app and marry it with ChatGPT there have been I don't know how many micro failures on this where we're like this is the shape it should look like and then throw that in slack and there's like a 2,000 message thread about how stupid we are and it's like that this is the thing I love about OpenAI is people will just tell us that right there's there's no uh there's no holding back on like when we fail with product things internally it's why the external product has been pretty great is because it goes through these cycles of like

所以对我来说,其实我失败过相当多次,而有时候不过是某个时间点上一切恰好对齐——技能、热情、市场的时机对齐了;拿这个项目来说,把我们在 Codex app 上学到的东西和 ChatGPT 结合起来,这中间不知道经历了多少次小失败,我们觉得"这就是它该有的形态",然后往 slack 里一扔,就冒出一条两千多条消息的帖子,讨论我们有多蠢;而这正是我爱 OpenAI 的地方——大家会直接把话摆给你,内部产品出问题、我们在产品上失败的时候,没人藏着掖着;对外的产品之所以相当不错,就是因为它经过了这样一轮又一轮的循环——

2000 this sucks

两千条,都在说"这东西真烂"。

I failed for like I don't know somewhere between 10 and 15 years before getting to this point so I'm still surprised every day that things are going well like and I know it I like but

在走到这一步之前,我大概失败了差不多 10 到 15 年,所以到现在我每天还是会惊讶事情居然进展顺利,我知道的,我……但是。

I think this is really important for people to that you can have a lot of things not work out and then things start to work out super well and it's just keep going and keep learning I imagine is is a lesson.

我觉得这对大家真的很重要:你可以有很多事都不顺,然后事情开始变得特别顺;我想,一直走下去、一直学习,大概就是那个教训。

Well, with that we're uh we reached our very exciting lightning round.

那么,说到这儿,我们就来到了非常激动人心的闪电轮。

I've got five questions for you.

我有五个问题要问你。

Great.

好啊。

Here we go.

我们开始。

Uh what are two or three books that you find yourself recommending most to other people?

有哪两三本书是你最常推荐给别人的?

See man, I'm I'm a parent now.

唉,伙计,我现在是当爸爸的人了。

I'm a parent of young kids, so I I'm like I don't know.

我有小孩,还小,所以我……我也说不好。

There's one called The Gruffalo to my kids.

有一本给我孩子读的,叫 The Gruffalo。

I Oh my god.

我……天哪。

Our kid just got obsessed with the Gruffalo.

我们家孩子最近就迷上了 The Gruffalo。

Yes.

是的。

Uh we have like a a bedtime chart and now it's like pajamas, brush teeth, books, Gruffalo uh on blankie.

我们有个睡前流程表,现在就是:穿睡衣、刷牙、看书、The Gruffalo、盖上小毯子。

Yeah.

对。

It's so good.

那本真好。

So other books I'm reading right now are like that style.

所以我现在在读的其他书也都是那种风格。

I'm like actually The Gruffalo is not a terrible one.

我觉得吧,The Gruffalo 其实还不算差的。

I feel like there's some less.

我感觉有些书更差一点。

So sweet.

太可爱了。

Yeah.

对。

Although every kids book is about death.

不过每本童书讲的都是死亡。

Like someone's eating someone, someone's killing.

就是谁吃了谁,谁杀了谁。

There's always like bad like murder.

总有那种坏事,像谋杀什么的。

Yeah.

对。

Yeah.

对。

And and destruction

还有毁灭。

like violence

像是暴力。

even when it doesn't feel like violence to them that like the words that there's like

就算对他们来说感觉不像暴力,但那些字眼里……

it creates like an arc and excitement.

它营造出一种起伏和张力。

Yeah.

对。

Okay.

好。

Great choice, Gruffalo.

选得好,The Gruffalo。

I currently have a book backlog.

我现在有一堆书还没读。

I I need like I need to read all of them.

我得把它们都读完。

Any other children's books that that your kid likes?

还有别的你孩子喜欢的童书吗?

Okay.

好。

So, yes, actually um I am well versed on children's space.

说真的,我对童书这块儿还挺在行的。

My favorite children's book ever is a very old one and it's called The Big Orange Splot or some something along those lines.

我最喜欢的一本童书是很老的一本,叫 The Big Orange Splot,大概是这么个名字。

Look it up.

你去搜一下。

It's great.

很棒。

Go get it.

去买一本。

Um if you hate HOAs, go get it.

如果你讨厌 HOA(业主协会),就去买一本。

It's about um this guy, Mr.

讲的是这么一个人,Mr.

Plumbean, who lives on a street where all the houses are the same.

Plumbean,他住在一条街上,那儿所有房子都一个样。

It's very neat street and then one day a bird drops a large can of orange orange paint on his house and he says, "F it." Like I'm going all in.

那是条特别整齐的街,然后有一天一只鸟把一大罐橙色油漆掉到了他房子上,他说:"去他的。"就是我要放开干了。

So he goes to the store, he gets paint, hammocks, alligators, and he like totally redo his house, and the neighbors like they're up in arms, right?

于是他去店里,买了油漆、吊床、短吻鳄,把自己的房子彻底翻新了一遍,邻居们都炸了锅。

Like, you know, HOA property value, whatever, whatever.

就是那套 HOA、房产价值之类的。

Um, it's not about HOAs, but I very anti-HOA person.

这书其实不是讲 HOA 的,但我是个非常反 HOA 的人。

It's just really cool.

它就是真的很酷。

And then like one by one, the neighbors go talk to them, have uh what they refer to as lemonade, but like it's a very convincing lemonade because one by one, they all start redoing their house.

然后邻居们一个接一个去找他聊,喝了书里所谓的"柠檬水",但那是一种很有说服力的柠檬水,因为他们一个接一个都开始翻新自己的房子了。

Like one guy does like a boat, right?

比如有个人把房子弄成了一条船。

I think I think that's a good book.

我觉得那是本好书。

I think people need to read this when they're especially, you know, newbies.

我觉得人们应该读读这本,尤其是刚起步的新手。

What I hear from this is agency.

我从中听到的是 agency。

Agency.

agency。

Exactly.

没错。

You can just do things.

你可以直接去做事。

Just do things.

直接去做事。

Amazing.

太棒了。

Okay, we'll keep going.

好,我们继续。

Uh favorite recent movie or TV show you really enjoyed if you've had any time.

如果你有空的话,最近有没有特别喜欢的电影或电视剧?

So, The Magic School Bus is back um on Netflix.

Magic School Bus 又回来了,在 Netflix 上。

It's It's a new animated.

是新的动画版。

It's got Kate McKinnon because Miss Frizzle is now Professor Frizzle and she's, you know, around but she's not the main Frizzle now.

里面有 Kate McKinnon,因为 Miss Frizzle 现在变成了 Professor Frizzle,她还在,但已经不是主角 Frizzle 了。

Kate McKinnon plays the main Miss Frizzle.

Kate McKinnon 演的是主角 Miss Frizzle。

Yeah, I always liked the magic school bus.

对,我一直很喜欢 Magic School Bus。

I've never used it.

我从来没看过。

Personally, like I don't have time for movies.

就我个人来说,我没时间看电影。

So, what I do is watch hour-long Netflix things back to back.

所以我的做法是一集接一集地看 Netflix 上那种一小时长的剧。

You know that thing that people do like

就是大家都会做的那种事,比如……

Oh, I couldn't sit I couldn't sit down and watch a whole movie and then like watch hour-long episodes and keep Yeah,

哦,我坐不住,我没法坐下来看完一整部电影,却能一集集地追一小时长的剧,一直看下去,对。

I did.

我就这样。

Some There's something addicting about that.

那种方式有种让人上瘾的东西。

Favorite product you've recently discovered that you really love.

最近发现的、你特别喜欢的产品是什么?

What was a terrible answer?

这算不算是个很烂的回答?

I feel like I'm discovering our product every day.

我感觉我每天都在发现我们自己的产品。

Beautiful.

漂亮。

That's so

这也太……

I think Linear does a great job.

我觉得 Linear 做得很棒。

Like linear until until this Linear was like my favorite at least software product.

在这之前,Linear 至少是我最喜欢的软件产品。

Is that what you guys used to to plan in?

你们以前就是用它来做规划的吗?

Well, in theory.

嗯,理论上是。

Do you have a favorite life motto that you find yourself coming back to in work or in life?

你有没有一句最喜欢的人生格言,是工作或生活中会反复想起的?

I want to ask everyone who works with me this because I I I feel like I'm not a motto person and then people tell me stuff I say all the time.

我想把这个问题问遍所有跟我共事的人,因为我总觉得自己不是个爱讲格言的人,可别人却会告诉我一堆我天天挂在嘴边的话。

Yeah.

对。

Like when we were chatting ahead of this, there's so many little nuggets that stood out to me that I've integrated into this chat.

就像我们录之前聊天时,有好多小金句让我印象深刻,我都揉进这次对话里了。

So I I totally hear that.

所以我完全懂你的意思。

Okay, last question.

好,最后一个问题。

You've been a PM, you've been a designer, you've been an engineer.

你当过 PM,当过设计师,也当过工程师。

Which is the toughest role of the three?

这三个角色里,哪个最难?

Which is like the hard to start a war?

哪个是那种一说就要吵起来的?

Yes.

是啊。

I know.

我懂。

I I think they're all very different.

我觉得它们都很不一样。

I the things that make it tough for one person make it easy for others.

让一个人觉得难的东西,恰恰让另一个人觉得容易。

There's a lot there's so much so many takes on this and like this triad.

关于这个、关于这个三角组合,有太多的看法了。

What's going to happen with this triad?

这个三角组合会走向何方?

Like designers are done or like should designers code?

比如设计师是不是要被淘汰了,或者设计师该不该写代码?

Are PMs cooked?

PM 是不是完蛋了?

Like are engine do we not need engineers anymore because PMs are going to write all the code?

比如工程师……我们是不是不再需要工程师了,因为 PM 会把所有代码都写了?

Or are designers going to PM now?

还是说设计师现在要来做 PM 了?

And like everybody's cooked and everybody's so back and I don't know like there's some convergence.

好像每个人都完蛋了,又每个人都强势回归,我也说不好,反正是有某种融合。

There's some fluidity that is being introduced that I think is refreshing and great especially for people with agency that want to be able to just like do the thing that needs to get done.

有某种流动性正在被带进来,我觉得这挺让人耳目一新、挺好的,尤其对那些有 agency、想着能直接把该做的事做成的人来说。

Um and at the same time like we talked about there are some things that shouldn't go away but I think people should find the stuff that's worth working on and go figure out what to do on those things.

同时呢,就像我们聊过的,有些东西不该消失,但我觉得人们应该去找到那些值得做的事,然后弄清楚在这些事上该怎么做。

That is a beautiful way to end it.

这个结尾真漂亮。

Andrew, thank you so much for being here.

Andrew,非常感谢你来。

Thank you.

谢谢你。

Bye everyone.

各位再见。

Thank you so much for listening.

非常感谢你们的收听。

If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app.

如果你觉得有价值,可以在 Apple Podcasts、Spotify 或你喜欢的播客应用上订阅本节目。

Also, please consider giving us a rating or leaving a review as that really helps other listeners find the podcast.

也请考虑给我们打个分或留个评价,这对其他听众发现这档播客真的很有帮助。

You can find all past episodes or learn more about the show at lennyspodcast.com.

你可以在 lennyspodcast.com 找到所有往期节目,或了解更多关于本节目的信息。

See you in the next episode.

下期节目见。

I love that thing about Brent in the Premiere.

我特别喜欢 Premiere 里关于 Brent 的那段。

Yeah,

是啊,

that's cool.

挺酷的。

I've actually used edit edits as well.

我其实也用它来做剪辑。

It's basically

基本上就是

simple stuff like, oh, could you just cut this into the three breaks, you know, like if there's a pause in conversation like the second and Codex understands it.

一些简单的事,比如,你能不能把这段切成三个断点,就是对话里有停顿的地方,像第二处那样,Codex 就能理解。

Basically, every job we feel like starts with a story like this, which is the product's not designed for it, but it's sort of a blank chatbot that can write code.

基本上,我们觉得每个任务都是从这样一个故事开始的:产品本来不是为它设计的,但它就是个空白的、能写代码的聊天机器人。

So, it can do everything, but like what

所以它什么都能做,但问题是

what are the useful things for it to do,

让它做哪些事才有用,

right?

对吧?

I mean, people just have to have curiosity about the project.

人们必须对这个项目有好奇心。

They they have to,

他们必须,

you know, have an in an intentional outcome and use Codex as the platform for that just to see what happens,

得心里有一个想要的结果,然后把 Codex 当成实现它的平台,就看看会发生什么,

you know, because there's no risk at all,

因为根本没有任何风险,

right? just a few tokens.

对吧?就几个 token。

Few tokens.

几个 token。

But of course, if you're working for OpenAI, there's less risks on that regard.

当然,如果你是给 OpenAI 干活,这方面的风险就更小了。

Yeah.

是啊。

You asked um

你问过

one of them like what or a lot of them actually like what skills are important and like then then you've also had conversations about like the cracked new grad versus the

其中一个人,或者说其实问过很多人,什么技能重要,然后你也聊过那种极强的应届生对比

like

就是说,

I don't know if you're married to the exact process you have right now like that.

我不知道你是不是死守着你现在的这套流程。

I don't know what advice to ever give, but if there's one piece, it's like do not get married to your exact process.

我从来不知道该给什么建议,但如果只有一条,那就是:别死守你现在的这套流程。

Get married to like the outcomes that you are uniquely able to deliver

要死守的是那些只有你能独一无二交付的结果,

and then do things like change your process to try things.

然后为了尝试新东西去改变你的流程之类的。

Like

就是,

you just keep

你就一直保持

I'm the best at understanding Figma auto layout like

我是最懂 Figma auto layout 的,类似这种

right

对

what are you doing,

你在干嘛啊,

right?

对吧?

Cuz AI is going to be better at that also.

因为 AI 在那方面也会比你更强。

You just keep spouting interesting things.

你就一直不停地冒出有意思的东西。

Keep it in.

这段留着。

It's crazy the level of self-awareness that's required to like be successful with AI.

要想靠 AI 做成事所需要的那种自我觉察程度,真的挺疯狂的。

It is.

确实。

It's also why I'm nervous to ever say like this is how something's going to be cuz I think about like my parents are open-minded people.

这也是为什么我总是不敢说“某件事就该是这样”,因为我会想到,我父母是很开明的人。

They are into their careers but just like the stuff that works here is just not going to it's not going to work with everybody.

他们在自己的事业上很投入,但在这儿管用的那套东西,就是不会,它不会对每个人都管用。

There's no nice way of saying that,

这话没有什么好听的说法,

you know, but like the people who are here are self- selecting for like, oh, I'm somebody who just figures out the next thing to do and that's just not like most of the population will not be an early adopter of things.

但在这儿的人是自我筛选出来的那类:哦,我就是那种自己会琢磨出下一步该干嘛的人,而这可不像大多数人——大多数人不会是新事物的早期采用者。

And there's also just like it's kind of like a a bummer to have to relearn things all the time.

而且还有一点,总是不停地重新学东西,其实挺让人扫兴的。

Yeah.

对。

You know, it's like god damn it

你懂的,那种感觉就是“真该死”

just to learn a new thing again.

又得重新学个新玩意儿。

Yeah.

对。

Like I I hate repetition.

我特别讨厌重复。

This is like a me thing.

这是我个人的毛病。

I don't.

我不。

This is why I'm like not ever the best media person either cuz I just like I hate repeating myself.

这也是为什么我也从来当不好媒体人物,因为我就是讨厌重复自己。

Perfect.

完美。

Um sucks to be a founder when you hate repeating yourself cuz you you like have to

讨厌重复自己的人当创始人就很惨,因为你必须得

repeater and chief.

首席复读官。

Um and so like to me I'm like if I can come in and do my job in a different way every day.

所以对我来说,如果我每天都能用不同的方式来做我的工作,

Love it.

那我爱死了。

But like that's not

但那可不是

You found product market fit for your job.

你为你的工作找到了产品市场契合(PMF)。

Yes.

对。

Yeah.

是啊。

No, that's how that's I like can't negotiate because I'm like well I don't want other jobs.

不,这就是为什么我没法讨价还价,因为我会想,我又不想要别的工作。

Amazing, man.

太棒了,兄弟。