Block如何成为全球最具AI原生特质的企业 | Dhanji R. Prasanna

它到底有多智能？

How how good is it?

如果按0到100分来评估，现在处于什么水平？就是那种你只需要思考和说话，它就能完成所有工作的程度。

Like, where is it at if you had to go zero to 100 of like, okay, it's it's gonna all you have to do is now think and talk and that'll just do your job.

没错。

Yeah.

语音交互是另一个重要组成部分。

So voice is the other big part of it.

所以它具备语音处理能力。

So it has voice processing capability.

所以它一直在监听他说的话，并试图进行解读。

So it's always listening to what he's saying as well and and trying to interpret that.

我认为这主要是个实验项目，毕竟他是我们Goose团队的核心成员，日常工作就是为Goose做贡献。

I would say that this is mostly an experiment, you know, given that he he's on our core Goose team and he contributes to Goose, so he has a day job.

这是他业余时间开发的一个项目。

This is a kind of thing on the side that he was developing.

等这个功能发展成Goose本身的原生功能，或是我们企业使用的其他工具时，我觉得它会大有可为，不过现在已经相当不错了。

So once this evolves into more more of a native feature of Goose itself or other tools that we use in the enterprise, I think it can have a lot of legs, but it's already pretty good.

我的意思是，这很可能帮他省去了大量繁琐的工作。

I mean, it's probably cutting down enormous amounts of busy work that he has to do.

举个例子，他会说'我有个会议冲突'。

So for example, one thing he'll do is he'll say, oh, I have a meeting conflict.

'那个时间我去不了'或者'我得去接孩子'。

I can't make it that time or I have to go pick up my kid.

然后Goose就会自动重新安排会议，完全不需要他坐在日历前反复点击十几次。

And Goose will automatically reschedule that meeting without him ever sort of sitting in front of his calendar and clicking through 10 times.

是的。

Yeah.

这些功能我们原本指望日历服务商来开发，但现在不需要了，因为AI能帮我们协调这些事。

So these are things that I think we were waiting for the calendar vendor to build as features into calendar, but we don't need to do that anymore because AI is able to orchestrate this for us.

这不是那个同时在四家初创公司兼职，结果把所有工作都搞得一团糟的家伙吗？

This isn't that guy that had, like, four jobs at four different startups that he was able to paralyze all his work in

不。

No.

不是的。

It's not.

他...他是我长期共事的伙伴，在Block公司工作多年。

He's he's he's someone that I've worked with for a long time, and he's been at Block for a long time.

他...他非常热爱实验，这种实验精神就像我们的Goose创始人一样，他们做着相同的事。

And he's he's just loves experimenting, and he embodies that culture of experimentation just like our creator of Goose who who did the same thing.

让我稍微展开说说这个话题。

So let me pull that thread a little bit.

你...你似乎已经瞥见了未来发展的方向。

You're you're kind of seeing a glimpse of where things are going.

Block公司在很多方面都遥遥领先。

You're very ahead of the curve in a lot of ways at Block.

你认为几年后工程师的工作方式、产品团队的工作方式会与现在有什么不同？

Where how do you think things will look in a couple years in terms of how engineers work, how product teams work that's different from today?

我认为很大程度上取决于LLM性能的提升，但我可以告诉你我正在尝试改变自己的工作方式，以及我们团队的工作方式。

I think a lot of it is dependent on the improvement of LLM performance, but I can tell you the way I'm trying to change how I work and how I'm trying to change our immediate team's way of working.

我觉得氛围编码是件有趣又激动人心的事——本质上就是和聊天机器人对话，让它帮你构建软件。

So I think vibe coding has been an interesting, exciting thing, which is you talk to a chatbot essentially and it goes and builds software for you.

但我认为这有很大的局限性。

But I think this is highly limiting.

这就像打乒乓球一样来回反复。

It's very ping pong.

比如，你做了某件事，等待三四分钟后，它返回的结果可能半生不熟，这时你需要不断调整、引导和完善，才能让它达到理想状态。

Like, you do something, you wait for three or four minutes and it comes back with something sort of half baked and you have to nudge it and guide it and massage it to get where it needs to be.

我认为我们将看到更多的自主性。

I think that we're gonna see much more autonomy.

我们正在与Goose合作进行几项实验，针对其下一版本，我们正努力推动它不仅能持续工作两三分钟，而是更长时间。

So we're working on a couple of experiments with Goose, with the next version of Goose, where we're really trying to push it to work not just for two or three or five minutes at a time.

我们的平均会话时长中位数是五分钟，平均七分钟，但我们正努力将其延长至数小时。

Our average, our median session length is five minutes and on average seven, but we're trying to push it to hours.

要知道，我们试图说明的是，所有这些大语言模型在夜间和周末人类不在时都处于闲置状态。

You know, we're trying to say, hey, all these LLMs are sitting idle overnight and on weekends while humans aren't there.

这完全没有必要。

Like, there's no need for that.

它们应该一直在工作。

They should be working all the time.

它们应该尝试预判我们的需求，就像回到对话前半部分那样。

They should be trying to build in anticipation of what we want if we go back to the earlier part of the conversation.

但我也认为，它们应该能够以过去无法实现的方式进行构建。

But, also, I think that they should be able to build in ways that were never possible before.

以前我们人类资源有限，带宽有限，协调成本高，因此不得不选择实验中的最佳路径。

Before we as humans, we had limited resources, limited bandwidth, and a lot of coordination overhead, so we would have to choose the best path to try in an experiment.

我认为现在我们不再需要这样做了。

And I don't think we need that anymore.

我们需要的是能够详细描述多个不同实验，然后或许睡一觉，第二天早上所有这些实验都已完成，我们可以直接淘汰其中五六个。

We need instead to be able to describe multiple different experiments in a great amount of detail, and then maybe we go to sleep and then in the morning, all those experiments are built and we can sort of throw away five or six of them.

我日常工作中有一项内容就是每天写代码，但更常见的是直接删除大量大量的代码。

So one of the things that I do regularly so I write code every day, but one of the things that I do regularly is just throw away huge, huge amounts of code.

这对我来说有点困难，因为我以前从未这样做过。

And it's kinda hard for me because I've never done that before.

我是说，我当然喜欢删除代码，但这次情况不同。

I mean, obviously, love deleting code, but this is different.

当你构建一个全新系统或功能时，突然觉得不太对劲，就会直接删除重头再来。

You build a whole new system or a whole new feature and you're like, oh, it doesn't feel exactly right, I'm just gonna delete and start over.

所以我认为你会看到更多这种工作方式。

So I think you're gonna see a lot more of that way of working.

而且我认为你会看到，与其重构应用来改变UI或升级版本，我们更倾向于直接重写整个应用。

And I think that you're gonna see, instead of us, for example, refactoring an app to have a different UI or to evolve into its new version, we're just gonna rewrite that app from scratch.

我一直在推动团队思考的问题是：如果每次发布都像RM减RF那样删除整个应用并重写，我们的世界会变成什么样？

And one of the things I'm really pushing our teams to think about is, what would our world look like if every single release, we RM minus RF, like deleted the entire app and rebuilt it from scratch.

虽然目前还做不到，但这展示了未来可能的方向以及这些工具将引领我们去往何处。

And so we can't really do that today, but I think this shows you some of the direction of what's possible and where these tools are taking us.

有趣的是，软件工程和产品领域有个通用原则：永远不要直接重写。

What's interesting about that is that there's kind of this common rule in software engineering and just product, don't ever just rewrite.

不要试图重写你的产品，因为你会遗忘多年来人们做出的所有小改进、调整和错误修复，你以为会很简单直接，结果却要花一年甚至更长时间才能回到原来的水平。

Don't try to rewrite your thing because you're gonna forget all of the small improvements and tweaks and bug fixes people have made over the years, and you think it's gonna be the simple straightforward thing and ends up being, now it's like a year or more of just getting you back to where it was.

现在AI居然能让这种操作成为可能，真是耐人寻味。

So interesting that AI now can make that possible.

你的意思是这实际上可能应该成为标准工作方式？

What you're saying is that's actually maybe the way you should be working.

我也这么认为。

I think so.

我觉得关键在于要让AI重视所有这些渐进式的改进。

And I think that the trick is getting the AI to respect all of those incremental improvements.

对，某种程度上，就像把这些改进‘烘焙’进去一样。

Yeah, and sort of, like, bake those in as Yeah.

作为规范的一部分，如果你愿意这么理解的话。

As a part of the specification, if you will.

是啊。

Yeah.

还有你提到的观点，这种智能体你只需给它一堆想法，它就能在一夜之间构建出来，我猜它还能进一步深入系统架构，自主产生创意并开始实施。

Also, the point you made about this agent just kind of you give it a bunch of ideas that builds them overnight, and then you could see I imagine it it goes even further up the stack and comes up with the ideas and starts building them.

然后你就会觉得‘哇，这主意太棒了’。

And then you're like, okay, oh, was a great idea.

现在我能在完全相同的工作流中立即看到效果。

Now I can see it immediately in the same same workflow.

没错。

Yeah.

确实如此。

That's that's true.

实际上我上周就在尝试你说的这种方法。

I was actually literally trying what you're saying just last week.

所以我们正在开发这个新版本的Goose。

And so I have this new version of goose that we're working on.

我要求它提出自我改进的想法并在一夜之间实施。

And I was asking it to come up with ideas to improve itself and implement it overnight.

有时它会

And sometimes it's

纸张滑动问题。

paper slip problem.

是啊。

Yeah.

有时它会完全偏离脚本，你得稍微往回拉一拉。

Sometimes it kinda goes off off the script entirely, and you have to sort of pull it back a bit.

所以我认为我们还没完全进入那个它能完全自我改进和自主的时代，但我确实觉得我们正处于一个过渡阶段——可以给它一点推力，比如列出10项我希望它能实现的功能清单，让它自己去摸索最佳实现方式。

So I think we're we're not quite at that era where it's completely self improving and completely autonomous, but I do think we're in a transition phase where we can give it that nudge and say, hey, here's my wish list of 10 things that I wish you could do, go and figure out the best way to do them.

如果功能描述足够清晰，它大约能成功完成60%的任务，剩下40%则需要人工干预和调整。

And it's successful, I would say, on, like, 60% of those things if the features are well enough described, and it struggles on the remaining 40 where you have to kinda intervene and and massage it.

对。

Yeah.

天啊。

Oh, man.

我已经在想象这样的未来：你给它‘驱动收入增长’的目标，然后它就直接说‘好的’。

I'm just imagining this future where you give it the goal of drive revenue and growth, and then it's just like, okay.

所有人被解雇。

Everyone's fired.

这是你们的薪水支票。

Here's pay here's your paychecks.

这里交给我吧。

I'll take it from here.

只是

It's just

我觉得我们不会去那里。

I don't think we're gonna be there.

说实话，我确实认为我们需要大量的人类审美来锚定这些AI，防止它们偏离剧本。

I I do think we're gonna need a lot of human taste to anchor these AIs so they don't go off script, to be honest.

这正是我们的设计负责人和设计团队推动我们思考的方向，我认为这将成为一个差异化优势，让我们超越当前人人都在谈论的AI泛滥时代。

And that's really where our our design lead and our design teams are pushing us to think, and and that's a differentiator that I think will push us beyond this era of AI slop that everyone's talking about.

所以，是的，这很像把它锚定在对人们重要的事物上——那些有品位、实用且有价值的东西。

So, yeah, it's very much like anchoring it into a thing that matters to people and the thing that's tasteful and useful and has value.

为了让这个更具体，有没有什么例子可以说明AI试图做某事或团队试图推销时，你不得不凭直觉判断？

To make that even more concrete, is is there an example of something maybe AI was trying to do or a team was trying to pitch where you had to just, like, know?

这就是人类需要介入并确保事情不偏离正轨的地方。

This is where humans are gonna step in and keep things keep things on track.

我会说更多是关于流程自动化这类事情。经常会有团队提出类似需求，说需要购买某供应商的新工具，因为现有工具无法实现X、Y、Z功能。

I'd say it was more around things like process automation or, you know, a lot of times I'll get this sort of request where a team will say, we need to buy this new tool from this vendor because our current tool isn't doing X, Y, and Z.

而另一个团队会说：不不不，我们可以直接用Goose构建一个应用，用一半甚至更少时间就能实现相同功能。

And another team will say, no, no, no, we can just use goose to build an app that will, you know, do the same thing for us in half the time, or less.

作为人类，这时你就会思考：这些真的有必要吗？

And then as a human, you're sitting there thinking, is any of this necessary?

比如，如果我们直接改变流程，是不是根本不需要考虑开发工具？

Like, if we just change the process, do we even need to think about building tools?

而这正是人工智能不擅长的领域。

And this is the thing that AI isn't good at.

它无法进行这种组合判断，也无法在全球范围内判断什么重要、什么无关紧要。

It's not able to have this sort of portfolio judgment or judgment across a global sense of what's important and what matters.

所以我经常告诉团队，要质疑基本假设，尤其是信息安全团队，因为他们有时会把自己绕进死胡同，拼命想保护某个东西。其实你可以让开发团队换种方式实现，或者如果那东西根本不重要，干脆就别开发。

So a lot of times, I tell teams, just question, like, the base assumption, particularly our infosec teams because they'll they'll twist themselves into knots sometimes trying to secure something, and you'll be like, well, ask the team that's building it to do it differently or to not build that at all if it doesn't matter.

这样你就不需要扩大安全防护的范围了。

And then you won't have to increase your surface area of securing it.

因此我认为这些领域更适合人类运用判断力，而人工智能目前表现不佳。

So I think those are the areas where it's better for a human to use judgment and AI has not done a great job.

你提到要自己开发软件工具，而不是购买现成的产品。

You make this point about building your own software, your own tools instead of buying stuff.

这是人工智能带来的一个重大问题。

This is a big question with AI.

它会取代所有这些SaaS应用乃至Salesforce吗？

Is it gonna replace all these SaaS apps to Salesforce over?

你们通过自主开发节省了多少成本？还是说反而对现有SaaS软件产生了新的敬意——尽管大家都在用且花费不菲？

Is there a sense of just either how much money you guys have maybe saved building your own stuff, or have you built a newfound respect for the existing SaaS software that everyone's using and and could pays lots of money for?

我认为企业容易陷入偏离核心使命的陷阱，而我们的核心使命是经济赋能。

I think there's a trap in getting away from your core purpose as a company, and our core purpose is economic empowerment.

即为客户、商家或艺术家提供销售、支付租金或上传最新创作到平台的能力。

So getting customers or merchants or artists the ability to make a sale or pay their rent or upload their latest creation to title.

我认为任何服务于这个目标的事物，我们都应该鼓励并投资。

And I think that anything that serves that purpose, we should encourage and we should invest in.

但如果纯粹从金钱角度衡量，那就偏离了我们的初衷。

But if we're just purely looking at dollars versus dollars, then that's, like, pulling us off that purpose.

比如，用自研工具替代供应商方案可能节省的成本，往往抵不上团队因此耗费的心力与技术专注力的损失。

Like, the savings and cost that there might be in replacing a vendor tool by something you build in house is probably not worth it in the mental bandwidth that you've lost and the amount of the team's sort of technical focus that's being taken away.

所以我认为应该始终回归公司最核心的价值，其他事情自然就会水到渠成。

So, I would say it's just keep coming back to the thing that matters to you as a company, and then the rest is, you know, we'll follow from that.

确实。

Yeah.

人们总是低估维护自建系统需要付出的持续成本——觉得'酷，我们一个周末就搞定了'，

I think people forget just how much maintenance it takes to keep something you've built like, okay, cool.

结果接下来数年都要面对无止境的维护、需求和支持工作。

We built it in a weekend, And now it's years of endless maintenance and requests and support.

正如你所说，这又回到了那个永恒准则：专注核心竞争力，其余皆可外购。

And and also to your point, it's like, it feels like it comes back to the always motto of just focus on your core competencies and then buy everything else.

没错。

Yeah.

这就是典型的二八定律问题，我们在为客户开发应用时已经深有体会。

It's the classic eighty twenty problem, and we have that enough with our with the apps that we build for our customers.

对吧？

You know?

比如我们开发过一些反响热烈的实验性功能，但后续不得不花大量时间解决长尾问题。

Like, we'll build some great experiments that that really resonate, and then we have to spend a lot of time ironing out the long tail of problems.

以Cash Card为例，我们基本上用一个周末——最多一周的整合时间——就完成了全部功能开发。

So in Cash Card, for example, we we built the entire functionality of Cash Card, I would say, pretty much in a weekend or maybe a week of sort of integration and work.

然后我们花了很长时间来解决这些边缘情况，比如有人会支付账单两倍的小费，这就会彻底破坏后端系统，或者人们把它当作加油站使用，而他们对信用卡的收费方式又完全不同。

And then it took a really long time to iron out all these edge cases where, you know, someone would tip twice the value of the bill, and then it would completely break something in the back end or, you know, people would use it as a gas station and they have a different way of billing your card.

所以，是的，很大程度上就是这样。按照你的观点，我总是会回到这个问题：我们做这件事的根本原因是什么？

So, yeah, it's very much that and and to your point, I would go always come back to like, what is the reason we're doing this?

为什么这对我们和客户来说很重要？

Why does it matter to us and to our customers?

如果它不能明确满足这一点，我就会把它当作无关紧要的事情推掉。

And if it doesn't clearly satisfy that, I would just push it off as a not interesting thing.

本节目由Perth Persona赞助，这是一个帮助组织用户注册、打击欺诈和建立信任的验证身份平台。

This episode is brought to you by Perth Persona, the verified identity platform helping organizations onboard users, fight fraud, and build trust.

我们在这个播客中经常讨论人工智能的惊人进步，但这可能是一把双刃剑。

We talk a lot on this podcast about the amazing advances in AI, but this can be a double edged sword.

每一个令人惊叹的时刻背后，都有欺诈者利用同样的技术制造混乱，洗钱、窃取员工身份和冒充企业。

For every wow moment, there are fraudsters using the same tech to wreak havoc, laundering money, taking over employee identities, and impersonating businesses.

Persona通过自动化用户、企业和员工验证来帮助应对这些威胁。

Persona helps combat these threats with automated user, business, and employee verification.

无论您是想防范候选人欺诈、满足年龄限制，还是保护平台安全，Persona都能根据您的具体需求帮助验证用户身份。

Whether you're looking to catch candidate fraud, meet age restrictions, or keep your platform safe, Persona helps you verify users in a way that's tailored to your specific needs.

最重要的是，Persona能让您轻松了解交易对象，同时不会给合规用户增加负担。

Best of all, Persona makes it easy to know who you're dealing with without adding friction for good users.

这就是为什么Etsy、LinkedIn、Square和Lyft等领先平台都信任Persona来保护他们的平台。

This is why leading platforms like Etsy, LinkedIn, Square, and Lyft trust Persona to secure their platform.

Persona还为我的听众提供每月500次免费服务，持续一整年。

Persona is also offering my listeners 500 free services per month for one full year.

只需访问persona.com/leni即可开始。

Just head to with persona.com/leni to get started.

就是persona.com/leni这个网址。

That's with persona.com/leni.

再次感谢Persona对本集节目的赞助。

Thanks again to Persona for sponsoring this episode.

关于AI讨论中最重要的话题之一就是人力资源招聘这类事情。

One of the biggest parts of the conversation around AI is head is hiring jobs, things like that.

所以我这里有个包含两部分的问题。

So there I have two kind of this two part question.

首先是这些AI工具的兴起和生产力提升如何影响了你们的人力规划和招聘方式？

One is just how has the rise of all these AI tools, this increased productivity, impacted the way you plan head counts and, hire?

其次，在AI成为你们工作重要组成部分的当下，你们现在招聘时会关注哪些与以往不同的特质？

And then what do you look for that's different in people you're hiring now that AI is such a big part of the way you guys work?

我认为目前技术发展程度还不足以从根本上改变构建像Cash App这种规模应用所需的人力数量。

I don't think that things have progressed far enough that it's really impacted in a fundamental way how would how many people you would need to sort of build an app of the scale of Cash App, for example.

对我们而言真正的变化完全不同，而且与AI毫无关系。

I think what's changed for us is much different and it has nothing to do with AI.

就像之前讨论过的，我们正在从GM架构转向职能型架构。

It's what we talked about earlier is moving from our GM structure to a functional structure.

在GM架构下，我们总是倾向于将工程人力视为商品。

And in our GM structure, our incentives were always to think of engineering headcount as a commodity.

因此如果我们想开发更多功能，就会简单地增加工程师数量，这就陷入了经典的人月神话陷阱。

And so we would just add more engineers if we wanted to build more features and the classic mythical man person month trap or whatever it's called.

我认为转向职能结构会彻底改变这一现状。

And I think that moving to a functional structure completely changes that.

你会觉得，我们可以利用通用平台和模块。

And you're like, well, we can leverage common platforms, common modules.

我们可以召集公司内外的专家，就如何更好地完成这项工作提供建议。

We can bring in experts from across the company to advise us on how better to do this.

因此我认为，这类变化使得我们的招聘方式大不相同，我们不再将工程师视为可以随意增加100人来开发Cash App下一款产品的商品。

And so those kinds of things, I think, have made it much different in how we hire, and we no longer see engineers as a commodity to just sort of add 100 people to go and build the next product in Cash App.

但在AI方面，我们非常欢迎那些积极拥抱这些工具并渴望尝试学习的人。

But on the AI side, we're very much looking for people that are embracing these tools and that are eager to try and learn from it.

我们并不要求应聘者一开始就是出色的AI实践者。

We're not looking for people who are amazing AI practitioners on the get go.

我想我们已经有这类人才，如果他们愿意加入我们，我们也很感兴趣。

I think we have those people and we're interested in those people if they ever wanna work with us.

但我更倾向于寻找那些真正渴望了解这些工具、持开放态度的应届毕业生，甚至是已经掌握这些工具的老手。

But I'm much more keen on looking for that college grad who just really is eager to learn about these tools and like open to it, or even the veteran who has embraced these tools and figured it out.

这就是我们在人才选择上的优化方向，而非特定技能组合。

And that's kind of where we're optimizing for who we look for rather than sort of a specific set of skills.

所以本质上，最大的变化就是寻找那些拥抱AI的人，而不是拒绝使用AI的人。

So essentially, the biggest change is just looking for people that are embracing AI, not being like, no, I don't need this stuff.

我是一名出色的工程师。

I'm an amazing engineer.

我不需要使用Cursor、Goose这些工具。

I don't need to use cursor or goose or all these things.

是的。

Yeah.

我会称之为学习型思维。

A learning mindset is how I would put it.

这是我们CEO杰克经常强调的理念——他希望我们成为一家学习至上的公司。

This is something that Jack, our CEO, talks about a lot, is he wants us to be a learning first company.

所以我们做的每件事、推出的每个实验，都要思考能从中学习到什么？

So everything we do, every experiment that we ship, what can we learn from it?

我们是否觉得自己已经全力以赴了？

And did we feel that we gave it our best shot?

我认为对他来说，这比每次都要得出正确的商业答案更重要。

And I think that that's more important to him than even sort of coming up with the right business answer every time.

那面试环节呢？

What about when you're interviewing?

你们会鼓励工程师在做练习题时使用AI工具吗？

Are you encouraging engineers to use AI tools as they're doing exercises?

过去一两年间这方面发生了哪些变化？

How does that how did that change over the past year or two?

是的。

Yeah.

我们现在正开始这样做。

We're starting to do that now.

传统上我们只用CoderPad之类的工具来白板解题，甚至用伪代码或类伪代码来编程。

So traditionally, we would just use, like, CoderPad or something like that to whiteboard through a problem or and even, like, program it in pseudo code or near pseudo code.

但现在我们在考虑的是，你能用Vibe代码构建些什么吗？

But now we're we're looking at, can you use Vibe code to build something?

你能说说你使用这些工具的熟练程度吗？或者你如何看待与它们共同发展？

Can you how are you how comfortable are you with these tools or how are you thinking about evolving with them as well?

但我得说现在还处于早期阶段，一个人是否精通Goose或Cursor这类工具，与他们是否是个优秀工程师之间，在我看来并没有必然联系。

But it's early days yet, I would say, that it's not clear to me that necessarily how someone knows how to use, you know, be it Goose or Cursor or any of these other tools matters that much to whether they're a good engineer.

我仍然认为我们过去面试看重的特质——批判性思维、深入理解问题技术本质的能力，远比是否是个完全AI原生的程序员重要得多。

I still think that things that we interviewed for in the past, a critical mindset, the ability to really understand deeply the technical nature of a problem is still much more important than whether you're a fully AI native programmer.

我一直思考的另一个问题是：哪个级别的工程师从这些工具中获益最多？这也是很多人的疑问。

Another question that I've always been thinking about a lot of people wonder is, what level of engineer is most benefiting from these tools?

有人会说初级工程师受益最大，因为他们现在能完成所有工作了。

You could argue it's the junior engineers, now they could just get all this work done.

也有人认为资深工程师获益更多，因为他们更懂系统原理，能调度成千上万的AI代理为其服务。

You could argue it's senior engineers because they know so much more about how things work and how they could just orchestrate thousands of agents doing their bidding.

你观察到哪个级别的工程师获益最多呢？

What have you seen in terms of which level is benefiting most?

是的。

Yeah.

对此我有两个看法。

So two answers to that.

首先你绝对正确——工程师资历越浅，他们就越愿意采用这些AI工具，也越容易上手。

One is you're definitely right that the more senior and the more junior they are, the more comfortable or the more eager they are to adopt these AI tools.

我认为这背后有多种原因，包括你刚才提到的那些。

And and I think that's for a variety of reasons, including some of them that you named.

我认为资深人士确实能深入理解一切运作的原理。

Like, I think the senior people really understand in great depths how everything works.

所以他们几乎松了口气，因为有这个工具可以代劳那些他们做过无数次却不愿再碰的事情。

And so they're almost relieved that this tool exists that can go and do all these things that they've done a million times before and couldn't be bothered.

而初级员工就像我侄女侄子摆弄黑莓手机那样。

And then the junior people are like my niece and nephew on a Blackberry or something.

他们飞快地处理事务——早年用黑莓，现在用iPhone。

They're just blitzing through thing not Blackberry in the early days and iPhones now.

当我还在键盘上笨拙摸索时，他们已瞬间发完短信。

They're blitzing through a text message when I'm still sort of seeking destroying through my keyboard.

可见我有多老派。

Shows you how old I am.

但最让我惊讶的是非技术人员使用AI代理和编程工具来构建东西，这实在令人惊叹。

So I think there's that, but I think the non technical people using AI agents and programming tools to build things is really what's been surprising and really amazing.

我认为这预示着未来这些角色将如何演变。

And I think that speaks to how these roles are gonna evolve in the future.

法律、风控甚至工程与设计之间的界限将会模糊。

The lines are gonna be blurred between whether you're in legal or in risk or in engineering and design even.

因此我认为，那些能善用工具优化自己工作流程的人，才是从中获益最大的群体。

And so I think that the people that are able to embrace it to optimize for their particular workday and their particular set of tasks are really, who are showing the most impact from these tools.

有趣的是，没人讨论工程生产力这个维度——它减少了公司其他部门对临时需求的频繁打扰。

It's interesting, no one talks about that element of engineering productivity, which is the reduction of asks from all the other parts of the company to build random one off things.

这对工程师来说无疑是巨大的效率提升。

That feels like a huge productivity gain for engineers.

规模非常庞大。

It is massive.

虽然我觉得这有点像修更宽的高速公路只会引来更多车辆的道理。

Although I think that it's a little bit like the analogy of if you build a bigger highway, you'll just get more cars on the road.

因此我认为，既然所有人都在开发软件，就意味着有更多软件需要开发、更多协调工作需要完成，每个人都更渴望加快交付速度并取得更大成果。

So I think the fact that everyone's building software means that there's more software to be built, more coordination to happen, and everyone's more eager to ship things faster and and with greater results.

所以我们看到整体开发速度的提升和对更多功能的需求，不知道这样说是否清楚。

And so we're just seeing an overall uptick in velocity and the ask for more features, if that makes sense.

是的。

Yeah.

完全同意。

Absolutely.

这与你提到的不放缓招聘的观点相呼应。

It connects to your point about you're not slowing hiring.

我听到的是，对更多工程师和产品人员的招聘需求完全没有减缓。

What I'm hearing is just headcount hiring desires for more engineers, more product people is not slowing at all.

你们基本上表现得就像AI根本不存在一样。

You're basically it's as if AI wasn't really there.

我们现在对此更加审慎了。

We're being more thoughtful about it.

就像我说的，在GM时代我们将其视为大宗商品。

So like I said, we were looking at it as a commodity in the GM era.

现在我们转为职能制，所需工程师数量与Square或Cash App的功能数量已不再直接挂钩。

And now that we're functional, it's much less about how many engineers we need as a function of the number of features we have in Square or Cash App.

在职能型组织结构中，我们更多考虑的是哪些领域可以优化、何处能建立深度，以及通过模块化、复用和深入平台开发等方式真正推动我们的优先事项。

And in the functional org structure, we think of it much more as what are the areas of optimization, where can we build depth, and what really accelerates our priorities through things like modularization, reuse, and going deep into platforms.

我特别赞同这个观点：如果想提高效率，别管什么人工智能，直接重组为职能型结构就行。

I love this hot take of, if you're trying to be more productive, forget AI, just reorg into a functional structure.

从某些角度看这确实没错。

It's not wrong in some ways.

这里还有个非常有趣的例子：我们正试图改进构建时间，同时使用Goose等众多工具来辅助实现这一目标。

So here's another really interesting example where we are trying to improve our build times and we're using you're using goose and a lot of other tools to help us with this too.

它们已经取得了显著成效。

And they've done remarkable things.

我们有个很酷的工具能分析测试套件，并根据代码变更选择最合适的测试来运行。

So we have this really cool tool that analyzes our test suites and selects the right test to run for changes that were made.

通过这种方式我们减少了约50%的测试运行，效果非常显著，而且避免了因不必要的测试消耗CPU资源而导致地球升温。

So we've cut down basically 50% of test runs this way, which is pretty great, and, like, we're not warming the planet as much with all these unnecessary CPU cycles being wasted on tests.

但将测试迁移到云端或直接删除过时的测试用例，可能还能再节省两三倍资源。

But then things like offloading tests to the cloud or simply just deleting tests that don't make sense anymore probably save you two to three times that.

因此，用个不太准确的说法，你仍然需要采取组合策略。

So there is still a portfolio approach that you need to take for lack of a better term.

就像我之前举的例子：该采购供应商工具还是自主开发？

It's like that example I told you earlier about, should we buy a vendor tool or should we build this in house?

其实更根本的问题是：这个流程是否真的有必要存在？

It's like, well, do we even need to do this process at all?

所以在某些方面，组织结构比工具的实际效能更重要。

So in some ways, structure matters more than the efficacy of the tools you have.

至理名言。

Wise words.

这让我想到埃隆·马斯克那套优化流程——其中关键一步就是：在开始优化和自动化之前，先问这东西是否真的有必要存在？

Makes me think about Elon has this whole process for how to optimize stuff, and one of the steps is, like, do we even need this thing before we start out optimizing and automating it?

在我展开讨论你职业生涯中的经验教训之前，你认为对于那些想更深入AI领域、或希望帮助团队更具前瞻性思维的人，还有什么特别有价值的建议吗？

Before I zoom out and ask about just general lessons that you've learned over the course of your career, is there anything else that you think might be really valuable or useful to folks that are trying to lean in further into AI or just help their teams think a little bit more forward thinking?

我会说一定要亲自尝试使用这些工具。

I would say really try and use these tools yourself.

我们推动技术落地的核心方式是：杰克用Goose，我用Goose，高管团队都定期使用Goose，同时也会使用其他AI编程工具和辅助系统。

So the way in which I think we've been able to drive most of the adoption is Jack uses Goose, I use Goose, our executive team all have used Goose, and use it regularly, and use other AI programming tools and assistance as well.

我们每天都在使用。

And we do it every single day.

通过亲身体验工作流程的变化，比读LinkedIn或《哈佛商业评论》的评论文章更能让你理解如何推动组织变革——毕竟实践出真知。

And so we learn a lot about how our own workflow can change, and that's gonna tell you so much more about how you're gonna change your organization's workflow than if you're reading a bunch of think pieces on LinkedIn or Harvard Business Review or whatever it is, and then trying to get your teams to follow suit.

其实我们对待所有事情都是这个原则。

So I think we do this with everything.

就是要亲身感受：自己先用产品，体会它的优缺点和人机交互，再思考如何推广给团队。

It's feel it, like use the product yourself, feel it, understand its strengths and weaknesses and its ergonomics, and then figure out how to apply it to your teams.

对此我有个非常认同的实用建议：别再光听我们空谈，停止纸上谈兵，直接动手构建些东西。

Something I found helpful in doing that, which I completely agree with, which is like, stop reading about it, stop listening to us talking about it, just like build some stuff.

确实如此。

The thing Yeah.

我发现的诀窍是：给自己设定具体任务或待解决的问题，这能提供真实动力让学习落地。

The the thing that I found really helpful there is have a specific task or problem you wanna solve for yourself, because that really motivates you and makes it very real.

比如就在前几天，我试图从谷歌文档里提取图片。

For example, just the other day, was trying to pull images out of a Google Doc.

你知道的，就像谷歌文档，我觉得它就像加州旅馆。

You know, like Google Doc, it's like, I think of it as Hotel California.

你把图片放进去，但除非做些疯狂的操作，否则根本弄不出来。

You put images in there, but there's no way to get them back out unless you do some crazy stuff.

所以我直接去Lovable建了个应用，只要输入谷歌文档链接就能轻松下载图片——搞定。

So I just went to lovable and like, build an app where I can give you a Google Doc URL and let me download the images really easily and bam.

对。

Yeah.

完美。

Perfect.

是啊，很好的例子。

Yeah, great example.

我几个月前也做过类似的事，当时我儿子有一大堆治疗单据。

I did something like this a couple months ago as well, where my son has a whole bunch of therapies.

他有特殊需求，我正试图收集所有治疗收据并分享给我妻子，她好向保险公司申报。这些单据格式乱七八糟——有的是截图，有的是PDF，搞得我焦头烂额。

He he has additional needs and so I was trying to gather all the receipts for all these therapies and share them with my wife and she she will like claim it from our insurer And I was really struggling to do this because they they're in various forms or screenshots in some cases or PDFs or whatever.

于是我让Goose处理，它发现能把这些收据全整理到苹果备忘录里，合并成一条笔记。

So I asked Goose to do this and it was all sitting on my laptop and Goose figured out that it could put all of these receipts into my Apple Notes app, into a single note.

它转成HTML格式后，就能无缝同步到我手机，之后我就能直接邮件或分享给我妻子。

It converted it to HTML, so it would sync seamlessly to my phone, and then I could email it or share it with her from there.

这种操作我压根想不到，它居然是用AppleScript实现的。

And that's just something I just never would have thought of, and it did this using AppleScript.

它就这样在后台帮我控制了我的电脑。

So it just controlled my computer for me in the background.

是啊。

And yeah.

所以这些工具以出人意料的方式帮助我们，而且正如你所说，越是用来解决实际问题，就越能理解它们的优势所在以及适用的场景。

So these are, like, surprising ways in which these tools help us, and the more you use them to solve real problems, to your point, the more you understand what their strengths are and where to where you can deploy them.

我超爱这个例子。

I love this example.

你是怎么做的？就直接去找Goose然后说‘这是我遇到的问题’吗？

How did so did you just go to Goose and be like, here's the problem I have.

你会怎么解决这个问题？

How would I how would you solve it?

对。

Yeah.

差不多吧。

Pretty much.

我说，我所有收据都在Google Drive里，所以我们遇到了相似的原始问题。

I said, I have all these receipts there in Google Drive, so we have similar origin problem there.

我需要把它们整理成统一格式，还需要核对总数完成这些操作。

And I need to get them into a single form, and I need to, like, collate the totals and do all this.

所以它先尝试了几种方法。

So it tried a few approaches first.

它尝试下载它们，用PDF阅读器来读取文件等等。

It tried to download them, and it tried to read them using a PDF reader and this and that.

关于Goose，我认为其他AI代理也能从我们这里学到的是，如果它尝试了几件事却失败了，它会退回来尝试不同的路径，并持续前进直到取得一些进展。

And then the thing about Goose that I think a lot of the other AI agents learn from us as well is if it tries a few things and fails, it'll back up and it'll try a different route, and it'll just keep going until it makes some prod some progress.

它确实做到了。

And that's what it did.

然后它选择了AppleScript作为实现方式，因为它有MCP扩展来控制我的电脑，这与我们前几天聊到的工程师用来监控屏幕的工具是相同的，但这次是针对一个非常具体的问题，它成功解决了。

And then it it picked AppleScript as a way to do it because it had the MCP extension to control my computer, and this is the same thing that our our engineer we were talking about the other day uses to watch his screen and things like that, but this was a very focused problem and it and it managed to do that.

所以，是的，这些工具的能力令人惊讶，而给予它们这种灵活性是学习如何使用它们的重要部分。

So, yeah, it's it's surprising what these tools can do and allowing them the kind of flexibility to do that is a big part of learning how to use them.

这很酷。

That is cool.

顺便问一下，普通人能运行Goose吗？能直接下载Goose来使用吗？

I love the by the way, can you run Goose like, as a regular person, can you just download Goose and use that instead of Yeah.

当然可以。

Absolutely.

是的。

Yeah.

没错。

Yeah.

你可以直接从我们的网址下载。

You can just download it from our our URL.

我们可以在节目备注里分享给你。

We can share it in the show notes for you.

对，你可以安装它。

And, yeah, you can install it.

我相信它支持Mac、Windows和Linux系统。

It it comes for Mac and Windows and Linux, I believe.

这是个Electron应用，所以能在所有平台上运行。

It's an electron app, so it'll work on all of them.

它还配备了命令行界面，为习惯使用命令行的用户提供了另一种操作方式。

It also has a command line, so for people who are more comfortable using that, we have that UI as well.

哇。

Wow.

你们确实是在与那些构建基础模型的大公司竞争啊。

You really are competing with these massive foundational model companies building.

用最简单的方式来说，Goose与其他产品相比有何不同？

Is what's the simplest way to compare Goose to something else?

它更像是Cloud Code这种最基础的类比，还是另有特点？

Is it like this Cloud Code, this simplest comparison or something else?

我认为它与Cloud Code有所不同，因为Goose本质上是一个实现MCP的平台。

I think it's a bit different than Cloud Code because at its core, Goose is a platform that implements MCPs.

MCP赋予了它动态可扩展的特性。

And so MCPs give it this dynamically extensible nature.

因此它能为你完成各种任务，无论是自动化处理（比如我们讨论过的Google文档和笔记这类事务），还是直接通过其他MCP执行编程任务，比如索引代码等操作。

So it can do all of these things for you, whether it's automating things like we were talking about with Google Docs and notes and things like that, or it can do straight up programming task for you using other MCPs, like so it can index code and do it that way.

所以它更像是一个可扩展的平台。

So it's really more of like an extensible platform.

我会说它介于传统AI助手（你只会问它今天天气如何那种）之间。

So I would say it sits somewhere between your classic AI assistant, where you just ask it, you know, what's the weather today?

你能计算一下从那个日期到现在已经过了多少个月吗？或者说，对于世界上那些更专注的光标和云代码来说，这算什么？

Can you calculate how many months it's been since this date or whatever it is to the more focused cursors and cloud codes of the world?

基本上，这就是一切的结合体。

Basically, it's everything combined.

天啊。

Holy.

而且是免费的。

And free.

你需要为LLM的token付费，不过...确实如此。

You pay for the LLM tokens, but but yeah.

是啊。

Yeah.

不像这些描述源模型那样...

There's not like this description source models with

哦，你...我的天啊。

Oh, you oh my god.

没错。

Yeah.

这太疯狂了。

This is crazy.

能参与构建Goose的团队真是太酷了。

What a cool team to be on building goose.

是Ed Block。

It's Ed Block.

老兄，肯定有

Man, must have

还有他们

an they

肯定玩得很开心。

must be having so much fun.

哦，天哪。

Oh, man.

好吧。

Okay.

让我把镜头拉远一点。

Let me zoom out a little bit.

你现在担任LinkedIn的CTO已经快两年了。

So you've been CTO of LinkedIn right now for just about two years.

在担任这个职位之前，你希望自己知道些什么？

What's something that you wish you knew before you stepped in this role?

如果能回到几年前，在你耳边悄悄说些小技巧、经验教训，你会说些什么？

If you could go back a couple years and just whisper a few tips and tricks or lessons and and see your ear, what would they be?

我想可能是两件不同的事。

I think maybe two different things.

一个是康威定律的力量，就像我们之前讨论的那样。

One is just the power of Conway's law, like we talked about before.

就是说在不改变组织中人员关系结构的情况下，想要改变结果有多困难。

It's like how difficult it is to change outcomes without changing the structure of relationships between people in an organization.

我觉得我潜意识里一直都知道这点，但能真正从内心深处领悟到它的重要性确实意义重大。

And I think I always kinda knew that at some level, but really appreciating it in a visceral way is big.

另一个我通过惨痛教训学到的道理是，人们往往只会在出问题时才发声。

The other thing that I really learned the hard way maybe is, you only hear about it when things are going wrong.

所以当一切顺利时，反而会陷入诡异的沉默，让你忍不住怀疑：我现在的方向真的对吗？

So when things are going well, you kind of have this eerie silence and you're like, well, am I doing the right things here?

我是否在解决正确的问题？

Am I focusing on the right problems?

因此保持判断力，定期抽身用全局视角审视事务——这些都是必须刻意安排时间坚持做的事，真希望我刚上任时就明白这点。

So having a bit of judgment, having a bit of time to step back and look at things holistically, those are things that you really need to make time for and and do on a regular basis, which I wish I had known when I took up the role.

回顾你在Block的时光——我总忍不住说成Square，毕竟这些年叫习惯了，但我知道Block才是集团总称，Square只是...这样解释大家能明白吗？

Looking back at your time at Block, I keep trying to I almost say Square because I'm so used to that over the year, but I know Block is is the name of the broader company, and Square is just one is that just so people understand?

Square是Block旗下的一个业务单元，是其产品线之一。

Square is one business unit, one product within Block.

没错。

Correct.

是的。

Yeah.

我们旗下有Square、Afterpay、Cash App和Tidal四大品牌，还有专注比特币业务的BitKey和Proto——这两个品牌负责硬件产品。

We have Square, Afterpay, Cash App, and Tidal are are four major brands, and then we also have BitKey and Proto that are focused on Bitcoin for us, and they we ship hardware in those two brands.

明白了。

Okay.

很好。

Great.

我觉得有些人会想，你们到底在说什么？

I think that some people are like, what are you even what are you guys talking about?

好吧。

Okay.

酷。

Cool.

回顾你在Block的时光，关于打造产品或组建团队，你学到的最反直觉、与大众认知相悖的经验是什么？比如那些违背常见创业智慧的观点？

So reflecting back on your time at Block, what's maybe the most counterintuitive lesson you've learned about building products or building teams that goes against what most people believe, say, common startup wisdom?

我认为代码质量是一个例子，作为一名工程师。

I think code quality is one, like, being an engineer.

我很早就明白了这一点，而且它一次又一次地被验证。

I learned this kind of very early on, and it it keeps coming true over and over and over again.

很多工程师认为代码质量对打造成功产品至关重要。

A lot of engineers think that code quality is important to building a successful product.

这两者其实毫无关联。

The two have nothing to do with each other.

我最喜欢的例子是YouTube。

My favorite example is YouTube.

YouTube被收购时我正在谷歌工作，我记得当时充斥着各种焦虑，说YouTube的代码库有多糟糕、架构有多差劲，他们居然把视频作为二进制大对象存在MySQL里等等。

I was working at Google around the time YouTube was acquired, and I just remember there was this whole wash of angst about how horrible the YouTube code base is and how terrible their architecture is and they're storing videos as blobs in MySQL and whatnot.

但你要知道，可以说YouTube是谷歌最成功的产品，远超其他，对吧？

And, you know, you could argue that YouTube is the most successful product at Google by a long way, right?

可能比他们其他许多产品加起来还要成功。

Like maybe more successful than many of their others combined.

因此，这与架构设计的好坏关系不大。

And so it really has very little to do with how well it was architected.

因为另一方面，谷歌视频——这个产品不知道大家是否还记得——它早在YouTube之前就存在了。

Because the flip side of that, Google video, which is product that I don't know if people remember, it existed before YouTube.

它支持更多格式。

It supported more formats.

它支持更高分辨率。

It supported higher resolution.

你可以上传长达一小时的视频。

You could upload you know, hour long videos.

YouTube当时完全没有这些功能。

YouTube had none of this.

它只有那种一两分钟的短视频功能。

It just had the, like, one or two minute quick video thing.

但它却以压倒性优势击败了所有竞争对手。

And it's far and away blown away its competition.

所以我认为要始终牢记核心问题：我们开发这些工具或应用程序的目的是什么？

And so I think just keeping that front and center is why are we building these tools or these apps?

或者说这些产品是为了帮助人们解决特定问题而存在的。

Or these products there for people to solve a specific problem.

就我们而言，它是为了让Square商户完成交易，比如向你售卖咖啡或手工制品。

So in our case, it's for a square merchant to make a sale, to sell coffee to you, or to sell something they've made.

这才是真正重要的。

And that's really what's important.

我们的Android平台性能如何并不重要，除非它能满足需求。

It's not really important how well our Android platform performs, unless it's serving that need.

因此我认为这是我职业生涯中非常棘手的问题，我不断遇到工程师们认为我们需要重构，需要用更好的方式来实现。

And so I think that's been a really hard one for me over my career, and I continually encounter engineers who think we need to refactor, we need to do this in a better way.

然后我就说，不，所有这些代码明天都可能被抛弃。

And then I'm like, no, this all this code could be thrown away tomorrow.

所以只需专注于我们正在构建什么以及为谁而构建。

So just focus on what we're trying to build and whom we're trying to build for.

这是一个非常深刻的见解和经验。

That is an incredible insight and lesson.

这个YouTube的故事真有趣。

This YouTube story is so fun.

而且是个很好的例子。

And such a good example.

你刚才说他们把视频内容存储在MySQL中，以行和列的形式作为blob数据。

You were saying they were storing the video, like content in a in a MySQL set, row and column as a blob data.

是的。

Yeah.

这就是我的意思，我实际上没有看代码，所以我无法验证，但

This is this is what I mean, I didn't actually look at the code, so I I couldn't verify it, but

这这就是那种

this this was the sort

普遍共识。

of common wisdom.

他们当时使用的完全是Python技术栈，与我们那时在谷歌高度优化的C++和Java服务器相比，速度慢得惊人。

And then they had a an entirely Python stack that was incredibly slow compared to the state of the art sort of c plus plus and Java servers that we had hyper optimized at Google back in those days.

这太滑稽了。

That is hilarious.

这让我想到公司内部的情况，当你在公司工作时，你会发现简直是一片混乱。

It makes me think about also companies, like, when you look inside a company, if you work at a company, you're just like, this is just pure chaos.

没人知道发生了什么。

No one knows what's going on.

整个体系随时可能崩溃。

This is just about to all fall apart.

而且基本上每个成功的超高速增长公司都是这样的状态。

And if that's basically what it's like at every successful hyper growth company.

确实，这话有一定道理。

Yeah, there's some truth to that for sure.

是啊。

Yeah.

所以我认为，对企业成功而言，有太多比代码更重要的因素，就像你说的：你是否在解决人们的实际问题？

And so I think, again, it's just, there's so much more that is more important to the success of a business, and it's what you said is, are you solving a real problem for people?

你能把产品送到用户手中吗？

Can you get in their hands?

你能持续为他们解决实际问题吗？

Can you continue solving real problems for them?

这与代码质量无关。

It's not about the quality of the code.

关键不在于内部运作得有多好。

It's not how well you operate internally.

完全同意。

Absolutely.

我认为在Cash App上，我们也遇到过这种情况。

I think on Cash App, we had that as well.

在Cash App早期，我从团队只有约10名工程师时就开始担任工程主管，直到团队发展到200多人，用户量也增长到了大约1000万到2000万左右。

So in the early days of Cash App, I was head of engineering from when we were about 10 engineers to, you know, 200 plus and took us to about 10 plus or 20,000,000 users thereabouts.

那里也有非常相似的情况。

And there was a very similar thing there.

从外部看，一切似乎都非常混乱。

It's like from the outside, it looked like everything was really chaotic.

就像人们会随意构建实验性功能并发布，看起来我们并没有严格遵守软件生命周期之类的严格政策。

It's like people would build random experiments and ship them, and it just didn't look like we were following strict policies on things like software life cycle and stuff like that.

某种程度上确实如此。

And it was kinda true.

我的理念始终是：我们拥有这么多才华横溢的工程师，如果试图把他们限制在非常严格的条条框框里，反而会弊大于利。

And my philosophy was always, we have all these brilliant engineers and I'm going to do more harm than good by trying to harness them into very strict sort of blinkered areas.

如果他们想花点时间做些完全是浪费时间的事情，但同时他们又在另一方面交付了惊人的成果，那我几乎会允许这种情况。

If they wanna spin their wheels building something that is a complete waste of time for a little bit, but at the same time, if they're delivering these amazing things on the flip side, then I'll almost allow that.

是的，我可以接受这种情况。

Like, I'll I'll be okay with that.

要知道这是个微妙的平衡，因为如果放任不管，工程师们真的会钻进死胡同里。

And, you know, it's a fine balance because engineers can really go off in into rabbit holes if you let them.

但是，没错，混乱确实能催生一定程度的创造力，你必须懂得如何在某些方面构建可控的混乱。

But, yeah, there's a certain amount of creativity that chaos breeds, and you have to know how to build controlled chaos in some ways.

所以你需要建立一个稳固的基础，避免破裂，就像避免重大责任问题那样，否则在我们这种情况下就会亏钱。

So you have to create a foundation that isn't, you know, liable to rupture, like you have major liability problems or something like that, or you're gonna lose money in our case.

只要这些基础稳固，并给予工程师们自由去实验、迭代和做那些让他们充满激情的事情，那就是理想状态。

And so as long as those things are bedded down and you allow your engineers to have the freedom to experiment and iterate and do the things that energizes them, like, that's the ideal.

说到可控的混乱，你在Block时期的头衔之一——我想这就是你为什么在Square待了四年半——是‘疯狂科学家’。

Speaking of controlled chaos, your one of your titles during your time at Block, what at at I guess this is why you're actually at Square was mad scientist for four four and a half years.

是的。

Yeah.

没错。

Yeah.

那段时间我主要是兼职工作，因为我有年幼的孩子需要额外照顾。

That was that was a time when I was working part time mostly because I had very young kids with lots of additional needs.

我当时是多个不同项目的顾问，试图帮助一些疯狂的想法落地。

And I was a consultant on various different projects, and I was trying to help sort of some wacky things get off the ground.

是的，我真的很感谢Block让我在职业生涯中拥有这样的自由角色。

And, yeah, I just I'm really grateful to Block that they afforded me the freedom to have that role in my career as well.

在进入‘失败角落’环节前——这个我稍后会解释——也许再问一个问题。

Maybe one more question before I take us to, to fail corner, which I'll explain.

你已经分享了一些职业生涯中学到的经验教训。

So you've shared a few lessons of things you've learned over the course of your career.

还有其他你认为对你工作成功至关重要的核心领导力经验吗？

Are there any other just, let's say, core leadership lessons that you've learned that you think have been important to you being successful at the work that you've done?

我认为凡事都要从小处着手。

I think start small with everything.

就像，如果你想煮沸整个海洋来泡一杯茶——我不知道是谁说的这句话，但它确实是个我常想起的实用比喻——你永远无法实现目标。

Like, if you try to boil the ocean to make a cup of tea, I don't know who said that, but it's a really, a useful phrase that I keep coming back to, You'll you'll never get there.

所以如果你要泡茶，就专注泡好那杯茶。

So if you're making a cup of tea, just make the cup of tea.

你不需要煮沸世界上所有的水。

You don't need to boil all the water that there is.

听起来那杯茶会非常难喝。

That sounds like really a not delicious tea.

海水泡的茶。

Ocean water.

是啊。

Yeah.

我记得卡尔·萨根还说过类似的话：如果你想从头开始做一个苹果派，就必须先创造整个宇宙。

I think there's another one of, like I think Carl Sagan said, if you wanna make an apple pie from scratch, you have to first invent the universe.

这就像是在说，要把注意力集中在眼前可实现的范围内。

So it's like, narrow your scope to the thing that's in front of you and that's achievable.

所以我认为这一点非常重要。

And so that that I think is really important.

这也是我们的核心理念之一，早在公司初创阶段就秉持着'从小做起'的原则。

And that's one of our core candidates and always has been even when we were just square in the early days, start small.

有没有什么成功或失败的具体案例？

Is there an example that that maybe worked really well or maybe didn't work?

是的。

Yeah.

我是说，Goose最初规模很小。

I mean, goose started small.

当时只是一位工程师利用业余时间，试图构建一个既实用又能满足其论文需求的项目。

It was just an engineer working on their own time trying to build something that was useful and that satisfied a thesis that they had.

所以，Goose的创始人Brad很早就深信这一点。

So, Brad, our creator of Goose, believed very early on.

我认为远在‘智能体将成为我们从大语言模型中获取价值的关键’这一流行词出现之前，他就构建了一个概念验证并分享给了许多人。

I think long before we heard the buzzword going around that agents would be how we unlock value from LLMs And he built a proof concept and he shared it with a bunch of people.

他将其分享给了Databricks和Anthropic。

He shared it with Databricks and Anthropic.

这让他们感到兴奋，同时也从他们那里学到了很多。

Got them excited and, you know, learned a lot from them.

于是项目就这样逐渐积累了发展势头。

And so it just sort of built momentum from there.

甚至在内部，情况也颇为相似。

And even internally, it was quite a quite a similar thing.

Cash App本身也是如此。

Cash App itself was like that.

Cash App最初差不多是黑客周的一个点子，后来逐渐发展成越来越大的项目。

And Cash App started more or less as a hack week sort of idea and grew into a bigger and bigger and bigger thing.

因此我们很多项目都始于这类小型实验，然后在此基础上不断扩展。

So a lot of our projects start with these small experiments that we try to then build on top of.

我们成为首家上市公司推出比特币产品，这其实又是一个黑客周创意，实际上是杰克、我和另一位工程师共同开发的。

We became the very first company that was a public company to launch a Bitcoin product, And that was, again, a hack week idea that, actually, Jack and me and another engineer worked on.

就是那个黑客马拉松团队，你和杰克·多尔西还有一位工程师。

That was the hackathon team, you and Jack Dorsey and an engineer.

对。

Yeah.

就我们三个人。

It was the three of us.

难以置信。

Unreal.

是啊。

And yeah.

而且那段经历很棒。

And and it was great.

当时我们去蓝瓶咖啡买了杯咖啡，是用比特币通过Cash Card支付的。

It was we we went and bought a cup of coffee at Blue Bottle, and it was bought using Bitcoin over Cash Card.

现在回想起来，那杯咖啡可能是史上最贵的一杯。

And I'll tell you those in in hindsight, probably the most expensive cup of coffee.

当时比特币什么价位？

What was Bitcoin at?

大概几千美元？

Like, thousand?

那时候是六七千美元。

6 or 7,000 back then.

是的。

Yeah.

哦，不。

Oh, no.

现在大约是12万了。

It's like a 120,000 now.

太好了。

Great.

对。

Yeah.

但确实，这是个例子，说明如果你先专注于一个小目标然后逐步构建，就能开发出对人们有用且可用的产品。

But yeah, it's an example of how, you know, you you get to a working useful product to people if you focus on a small thing first and then build.

再强调一下这一点，这与那种‘我们有个大想法，就要投入大量资源立即大干一场’的做法截然相反。

Just And to double down on this, this counter to, okay, we have a big idea, we're just gonna put a bunch of resources on it and go big immediately.

没错，完全同意。

Yeah, absolutely.

我也曾参与过那样的团队。

And I've been part of teams like that too.

在我的职业生涯中，我在谷歌参与过一个叫Google Wave的产品，它试图满足所有人的所有需求。当时我们甚至有七八十名工程师在开发这个东西，而它真正在谷歌以外的用户还几乎没有。

So in my career, I worked at Google on this product called Google Wave, which was trying to be everything to everyone, and, you know, we were seventy, eighty engineers building this thing before it even really had any users outside Google.

所以我认为这是个一开始就追求宏大、试图首日就一鸣惊人的例子，可能缺乏一些脚踏实地、根据现实情况调整的务实精神。

And so I think that's an example of something that started big, tried to go big on day one, and probably lacked some of that meeting the Earth where where reality lies and and and adapting accordingly.

我记得Google Wave。

I remember Google Wave.

完全正确。

Absolutely.

它很美。

It was beautiful.

炒作得很厉害。

A lot of hype.

我不记得具体是为了什么，但它看起来确实很棒。

I don't remember what it was for specifically, but it it looked really nice.

是啊。

Yeah.

我是说，那次经历让我学到了很多。

I mean, lot of learnings from that one for me.

没错。

Yep.

还有别的吗？

What else?

还有其他重要的经验教训吗？

Any other big lessons?

这两点是最主要的，但我也想补充说，要对所有基于假设的问题提出质疑。

Those two are the big ones, but I would also say, like, question based assumptions on everything.

你知道，有时候我们会陷入这样的陷阱：作为专业人士，我们过于专注于当天、当周、当月正在构建的东西。

You know, sometimes we get into traps where we are as professionals hyper focused on what we're building that day, that week, that month.

而我们没有停下来思考：我们真的应该构建这个吗？

And we don't stop to think, should we even build this at all?

建造这个的目的是什么？

Or what's the purpose of building this?

我们能否构建完全不同的东西，更符合我们存在的核心意义？

Could we build something completely different that would matter more to our core reason for being?

所以我会说，没错，要质疑那些基本假设。

So I would say, yeah, question the sort of base assumptions.

虽然有点老生常谈，但你确实需要不断提醒自己反复应用这一点。

It's somewhat of a cliche, but you really need to remind yourself to apply it over and over and over again.

我之前在播客上采访过你的一位同事IO，他曾和你一起开发Cash App。

I had a colleague of yours on the podcast back in the day, IO, who worked with you on Cash App.

是的。

Yeah.

他是我的朋友。

He's a friend of mine.

他非常出色。

He's amazing.

他有句名言大意是——我记不太清了——但意思是要触及你正在研究的事物的本质。

He's the quote along those lines of just like I forget exactly what it was, but it was just like, get to the bare metal of the thing that you're working on.

就是要亲手触碰你正在构建的东西，深入其基础才能真正理解发生了什么。

Just like touch the thing that you're building and go to the the base of it to really understand what's going on.

想象这在开发Cash App和Cash Card时非常重要。

And imagine that was really important with building Cash App and Cash Card.

是的。

Yeah.

IO是我共事过的最优秀的产品人之一。

IO's one of the best product people I've ever worked with.

而且，说实话，他也是我最亲密的朋友之一。

And, know, one of my closest friends actually.

所以绝对要支持他，你也应该支持他。

So absolutely with him on and you're on that one.

是的。

Yeah.

好的。

Okay.

接下来我要带大家进入播客的固定环节，我称之为'失败角'。

I'm gonna take us to a recurring segment on the podcast I call Fail Corner.

你已经分享了一个你参与过的失败产品案例。

You already shared one example of a product that failed that you worked on.

我想知道是否还有另一个例子。

I'm curious if there's another.

问题很简单：你参与过哪些没能成功的产品？

And the question is just, what's a product you worked on that did not work out?

因为听众们总是听到这些了不起的成功人士来播客分享各种成功故事，无尽的成功，但他们很少听到事情不如人意的故事。

Because people listening to this have all hear all these amazing successful people come on the podcast, share all these stories of success, endless success, but they don't hear the stories of when things don't work out.

所以这个问题就是：你参与过哪些失败的产品？它给你带来了什么教训？

And so this question is just what's a product you worked on that didn't work out, and what did that teach you?

这是个非常有价值的观点。

It's a very valuable point.

我是说，我的职业生涯基本上就是一连串失败产品叠加失败产品。

I mean, my career has basically been a string of failed product on top of failed product.

而且我觉得，比如Google Wave的例子，我在Google Plus上做过Hot Minute，那也是场史诗级的失败。

And I think that, yeah, the Google Wave examples there, I worked for Hot Minute on Google Plus, which was another epic failure.

说得好。

Good one.

我在一个叫Secret的社交网络初创公司工作过，它曾红极一时，但很快就爆炸式消亡了。

I worked at this social networking startup called Secret, which, you know, burned hot for a bright minute and then blew up.

后来我们又做了个电子邮件创业项目，同样看起来很有前景。

And then there was an email startup that we did, and that was, again, very promising.

结果还是逐渐凉了。

And then that sort of fizzled.

所以Canva的联合创始人跟我一起搞的那个项目。

So, the the co founder of Canva and I worked on that one.

总之就是一连串失败，但每次我都学到些东西，明白以后绝不能重蹈这类覆辙。

So, there's been a whole string of failures but at each point, I think I learned something and I learned that, you know, I need to never make that class of failures or errors again.

而Cash App大概是我最大的成功案例，这个早期参与的产品最终成长为人们喜爱的商业巨擘。

And so Cash App was probably like the big success for me that a product that I worked on that was very early on and grew to be this sort of giant business and product that people love.

所以我的职业生涯本质就是从失败中汲取教训，过程中也学会了谦逊，开始愿意倾听他人观点——特别是批评意见，不再自以为是。

And so, yeah, that's been my career is essentially taking the learnings from all these failures, getting some humility out of it in the process too, coming into things, willing to listen to other people's points of view, critical points of view, and not just kinda thinking that I have all the answers.

是啊。

Yeah.

我打赌这些失败产品的代码肯定很漂亮，架构设计也做了很多优秀决策。

And I I bet all these products that fail had really beautiful code, a lot of really good architecture decisions were made.

他们中的一些人，无论从哪方面看都很糟糕。

Some of them some of them were awful in every way.

它有太多失败的理由。

So many reasons for it to fail.

难以置信。

Incredible.

Dhanji，在我们进入激动人心的快问快答环节前，你还有什么想分享的，或者，不知道，想再次强调的吗？

Dhanji, is there anything else that you wanted to share or, I don't know, double down on before we get to our very exciting lightning round?

我想说，我们正处在一个充满变革的时代，人们对未来感到恐惧、沉默或不确定。

I would say, you know, I think that we're in this era of a lot of change, and people are scared or reticent or uncertain about where things are going.

我认为应该关注那些对你重要的事情。

And I think that look at the things that matter to you.

对我们来说，就是开源、开放协议，为每个人改善访问条件。

You know, for us, it's open source, open protocols, improving access for everyone.

我很幸运，职业生涯中只开发过免费或几乎免费的产品，或者有免费基础版，然后你可以付费升级一些高级服务，这些产品人人都能用。

You know, I've been very lucky in my career to only work on products that are either free or almost free to anyone, you know, or they have a free tier, and then you're kinda pay for some premium services and that are usable by everyone.