当代码变得廉价后的软件工程——对话萨提亚·纳德拉

Software Engineering After Code Becomes Cheap with Satya Nadella

萨提亚·纳德拉 Satya Nadella · The Pragmatic Engineer · 2026-11-16 · 约 64 分钟 · 原视频 ↗

打开互动全文版(中英对照 + 问答)→

本期速览 · Overview

微软 CEO 萨提亚·纳德拉谈 AI 时代真正稀缺的不是写代码,而是验证代码。

Microsoft CEO Satya Nadella on why verification, not code generation, becomes the scarce skill in the AI era.

要点 · TL;DR

核心观点 · Key points

反共识 · Contrarian takes

本期章节 · Chapters(共 28)

全文 · Full transcript(中英对照)

引言 Introduction

Host

在一个代码生成变得廉价的世界里,变得困难的是验证。事实上,我觉得我得学习,因为我还没有在一个主要以代码为输出的团队中工作过,但代码作为输出有一个特定的操作顺序。并不是说代码是输出就意味着每个人都在生成各种随机的代码。

In a world where code generation is becoming cheap, what is becoming hard is verification. In fact, I kind of feel like I got to learn because I've not worked in a team that is mostly now code as output, but the code as output has a particular order of operations. It's not like code is output means everybody is just generating all sorts of random code.

Host

不知从哪儿冒出来的,2025 年的时钟代码。我第一次感觉到你没有立即回应。

Out of nowhere, 2025 Claude Code. For the first time I felt that you didn't have an immediate response.

Host

当模型跳跃非线性地发生时,形态必须改变。从这个意义上说,AI 优先的产品构建是无情的。大约六个月后,IDE 就逐渐消失了。

When model jumps happen non-linearly, the form factor has to change. AI first product building in that sense is unforgiving. In six months or so, IDEs kind of faded away.

Host

我在问自己,开源运动和现在的开放模型之间有相似之处吗?

I'm asking myself, are there parallels to the open source movement and now the open model?

Host

如果你只是越过这个叫做“将有一个前沿模型拥有世界上所有知识”的东西,对吧?如果你不相信这一点,那么你必须假设在某种意义上,世界上应该有多少公司就有多少前沿模型,对吧?因为毕竟,

If you just get past this thing called there's going to be one frontier model to have all of the world's knowledge, right? If you don't buy that, then you have to assume that there should be in some sense as many frontier models as firms in the world, right? Because after all,

Host

在代码生成变得非常廉价之后,软件工程意味着什么?今天的嘉宾是 Satya Nadella,微软 CEO,少数几位以软件工程师身份开始职业生涯的大型科技公司 CEO 之一。今天我们讨论“软件工厂”这个词对 Satya 意味着什么,以及它今天在微软内部的样子,OpenAI 模型、本地模型,以及无计量智能的想法,我们开发者对 AI 非常真实的焦虑,微软的自愿离职提议,以及为什么初级招聘放缓可能只是暂时的,还有更多。如果你想了解 Satya 认为在这个 AI 时代成为一名出色的软件开发者需要什么,并想了解他关于如何理解所有这些变化的深思熟虑的心智模型,这集就是为你准备的。

What does software engineering mean after code generation becomes very cheap? Today's guest is Satya Nadella, Microsoft CEO, one of the few big tech CEOs who started his career as a software engineer. Today we cover what the term software factory means for Satya and how it looks like today inside of Microsoft, OpenAI models, local models, and the idea of unmetered intelligence, the very real anxiety for us devs around AI, Microsoft voluntary exit offers, and why the junior hiring slowdown might only be temporary, and many more. If you want to understand what Satya thinks will take to become a standout software dev in this age of AI and want to understand his thoughtful mental models on how to make sense of all this change, this episode is for you.

Host

本集由 Turbo Buffer 呈现,这是一个极其可扩展、快速且廉价的混合搜索引擎,构建在对象存储之上,被 Anthropic、Linear、Ramp 等前沿公司使用。许多切换到 Turbo Buffer 的公司将搜索费用削减了 50% 或更多,并获得更快的搜索结果。请在 turbobuffer.com/pragmatic 查看。

This episode is presented by Turbo Buffer, a ridiculously scalable, fast and cheap hybrid search engine built on top of object storage used by cutting-edge companies like Anthropic, Linear, Ramp, and many others. Many companies who switch to Turbo Buffer are cutting their search bills by 50% or more and getting much faster results. Check it out at turbobuffer.com/pragmatic.

Host

Satya,欢迎来到播客。

Satya, welcome to the podcast.

Satya

非常感谢你邀请我,Gurugay。

Thank you so much for having me, Gurugay.

Host

嗯,很高兴来到这里,因为我不确定你是否知道,但我曾是微软的 D。

Well, it's great to be here because I'm not sure if you know, but I was a Microsoft D.

Satya

你在这里还是在哪里?

Were you here or where were you?

Host

我当时在伦敦,这是我第一次来到总部。所以,你知道,大约 12 年前,但我想回顾一下。2014 年你成为 CEO 时我实际上在微软,我清楚地记得微软当时仍然非常拥抱开源,而当你上任时,你开始拥抱开源。Visual Studio Code 发布了。微软收购了 GitHub,并在我们使用 TFS 之前全力投入 Git。TypeScript 作为开源发布。我只是想回顾一下。为什么你们人们感到惊讶?我在微软以外的朋友感到惊讶。连我也感到惊讶。为什么微软全力投入开源?

I was in London and this is the first time I'm here in headquarters. So, you know, like 12 12 years back, but I want to rewind back there. I was actually at Microsoft when you became CEO in 2014 and I remember, distinctly remember how Microsoft was still very much embracing open source and when you came on, you started to embrace open source. There was Visual Studio Code that was released. Microsoft acquired GitHub and went all on Git before we used TFS. TypeScript was released as open source. And I just wanted to rewind back. Why did you people were surprised? My my friends outside of Microsoft were surprised. Even I was surprised. Why did Microsoft go all in on open source?

Satya

是的,我的意思是,在某种程度上,有时这些叙述和看法会被塑造,改变需要很多努力等等。因为如果你看看甚至,我会说甚至提交到 Linux 的代码,甚至在我成为 CEO 之前,我认为微软有重大贡献并致力于此。但我认为我们真正确保内部理解并且世界理解的事情是微软致力于开源,对吧?无论是将我们的一些工作负载(如 SQL Server)带到 Linux,还是为 Linux 做贡献,或者开源 TypeScript 等项目。我认为我们都是其中的一部分,而且你知道,显然我们作为 GitHub 的管理者让我们处于中心位置。我们只是想确保我们能够通过正确对待社区来赢得开源社区的信任。事实上,这很迷人。我一直思考这个问题,当你看看使用开源的公司和为开源做贡献的人。如果在某个时候有一个真实的统计,我认为微软会排在首位,或者至少名列前茅,对吧?事实上,如果你看看一些所谓的其他从开源中受益最多的公司,却几乎没有贡献。从来没有。我认为这就是,你知道,嘿,你不能抱怨,但世界就是这样运转的,总是叙述与事实的较量。

Yeah, I mean, at some level, sometimes these narratives and perceptions get shaped and it takes a lot to change and what have you. Because if if you looked at even, I would say commits even into Linux, even pre me becoming CEO, I think Microsoft had significant contributions and were committed to it. But the thing that I think we really made sure that we understood internally and the world understood is Microsoft is committed to open source, right? Whether it is bringing some of our workloads to Linux like SQL Server or contributing to Linux or or open sourcing projects like TypeScript and others. I think we're all part and you know, obviously our you know, us as stewards of GitHub put us in the center of it. And we just wanted to make sure that we were able to earn the trust of the open source community by doing right by the community. In fact, it's fascinating. I've always thought about this, which is when you look at companies that have used open source and people who have contributed to open source. If there was a true accounting of it at some point, I think Microsoft would be at the top or if not in amongst the top, right? In fact, if you look at some quote unquote a lot of these other companies that have benefited the most from open source, have hardly contributed. It's never. And that is, I think, how, you know, and like, hey, you can't complain, but that's how the world rolls, which is it's on narratives versus facts all the time.

Host

嗯,这是叙述与事实的较量。没错。而且,我认为 90 年代的 Windows 遗产改变了人们的看法,你知道,就像他们如何看待它。但我记得对开发者社区产生最大影响的并不是真正的开源。而是 Visual Studio Code。

Well, it's it's narratives versus facts. That's true. And And again, I think there's that Windows legacy from the '90s that shifts people's you know, like how they think about it. But what I remember made the biggest impact across the developer community was not really open source. It was Visual Studio Code.

Satya

是的。

Yeah.

Host

它确实是开源的,当然,但它也是免费的。在我使用 Visual Studio 之前,我是一名开发者,它太好了。Visual Studio 2003 仍然是一款了不起的软件。不知何故,你们决定微软将 Visual Studio 以免费的开源版本带给更广泛的开发者社区,然后砰,10 年或 5 年后,它无处不在。每个人都在使用它。这是有意的吗?

It was open source, sure, but it was also free. And I was a developer before I used Visual Studio, and it was so good. Visual Studio 2003 is still amazing piece of software. And somehow you decided Microsoft to bring Visual Studio for free in an open source version to the broader developer community, and boom, 10 or 5 years later, it's everywhere. Everyone is using it. Was that intentional?

Satya

是的,事实上,你知道,Eric Gamma 和其他当时在那个位置的人绝对是——这是有意的,甚至与社区一起构建。事实上,你知道,当我看着甚至产品制作时,那是我第一次接触到我认为现在人们谈论的,你知道,在开放中构建。那真的是我们所有人第一次学习在开放中构建意味着什么,拥有围绕它的社区,以及那个反馈循环。我记得当时甚至花时间与 Amanda Silver 等人一起学习那个团队,甚至产品经理——在那个团队做 PM 意味着什么,他们如何每天将社区反馈视为构建的方式,而不是,你知道,自上而下——你知道,就像不是,你知道,我听我的副总裁。我听我的社区。所以除了开源之外,在开放中构建有许多许多文化方面,我认为这些是根本性的变革。事实上,我会说有时我有点觉得我们在那里学到了很多伟大的教训。我们应该更深入、更广泛地学习它。

Yeah, in fact, you know, that, you know, Eric Gamma and others who were at that time in that place really was absolutely It was an intentional Even building with community. In fact, you know, when I look at even product making, it was the first time I also got exposed to what I think now people talk about, you know, building in the open. That was really the first time we all learned what building in the open means, having the community around it, and just that feedback cycle. I remember even spending time at that time with Amanda Silver and others learning how that team, even the product manager What does it mean to be a PM on that team, how they every day looked at the community feedback as a way to build versus, you know, top You know, like it's not like You know, I listen to my vice president. I listen to my community. So there were many, many cultural aspects of building in the open beyond open source that I think were fundamentally transformative. In fact, I would say sometimes I sort of feel like we learned a lot of great lessons there. We should have learned it even more deeply and more broadly.

Host

说到更深入或更广泛地学习,Visual Studio Code 无处不在。GitHub 做得很好。然后有了 Copilot,我的意思是,即使回顾过去,我也非常惊讶微软如此早地意识到 AI,在 ChatGPT 3.5 之前,我认为这是 3.0 版本左右,Copilot 像魔法一样工作,至少有一年没有真正的竞争,直到 2023 年 Cursor 出现。

Speaking of of learning more more deeply or more broadly, Visual Studio Code was everywhere. GitHub was doing great. And then with Copilot, I mean, I was so surprised even looking back, how Microsoft was so early to realizing that AI, before ChatGPT 3.5, this was I think the 3.0 version or or so, Copilot was working like magic and there was no real competition for at least a year until 2023 when Cursor came out.

CLI转型与IDE的未来 The Shift to CLI and the Future of IDEs

Host

然后 Cursor 出现了,我仍然觉得你们在 VS Code 上非常聪明,因为 Cursor 是基于 VS Code 的,界面熟悉,而且你知道,GitHub Copilot 和 Cursor 在并驾齐驱地竞争,互相追赶。然后不知从哪儿冒出来,2025 年的 Claude Code。对我自己来说,对在初创公司工作的人,甚至大科技公司的人来说,6 个月内大家就都忘了 IDE。我仍然非常惊讶,微软,你们呼吸着开发者 DNA,它一直在你们的 DNA 里。你怎么看你们错过了这个突然的转变,而且我第一次感觉到你们没有立即回应?

And then Cursor came out and I still felt that you guys were super smart about VS Code because Cursor was based on VS Code, familiar interface and you know, GitHub Copilot and Cursor were going side by side competing with one another, keeping up with one another. And then out of nowhere, 2025 Claude Code. And for myself, for people working at startups, even big tech companies, in 6 months everyone just forgot about the IDE. And I am still so surprised that Microsoft, you guys live and breathe, developer DNA is in your DNA and it's always been. How do you think you missed that this sudden shift would come and it for the first time I felt that you didn't have an immediate response?

Satya

是的,这是一个值得展开的好话题,对吧?我想正如你所说,早早进入这个领域,当时我们还在 GPT-4 之前,对吧?所以当我们甚至刚开始的时候,代码补全就已经很神奇了。那真的是 GitHub Copilot 的雏形,如果你还记得我们一直在做 IntelliSense,我们取得了不错的进展,然后我们说哇,这东西能行。坦率地说,当时有很多人怀疑它是否能成功,但它开始了,而且当我开始看到这一点时,我自己对缩放定律的信念从根本上得到了 reaffirmed。我认为我们在构建那个产品方面做得不错。用模型构建的关键之一是,当模型出现非线性跃升时,产品形态必须快速改变,对吧?这就是所谓的 AI 优先产品构建,在那种意义上是很无情的,因为你基本上必须围绕给定的模型构建一切,然后为下一个模型可能具有的涌现能力做好准备,并为此重新调整自己,对吧?不要爱上你构建的东西。所以这就是为什么即使当智能体式编码或任务处理开始成为可能时,对吧?不仅仅是代码补全,而是你可以分配小任务。那就是 Cursor 真正做得很好的地方,你知道,他们拿走了 IDE,他们分叉了,但他们在智能体模式上做得很好。然后我们在智能体模式上赶上了,因为那是我们甚至可以在 OpenAI 模型之上做的事情。但然后击中我们的事情显然是 Anthropic 在他们的模型上做的出色工作,这真正使 CLI 成为一种产品形态,然后你甚至可以说只是完成代码的完全处理,有效地从输入到输出,对吧?我的意思是,我认为那就是大约 6 个月内的转变。我们主要转向了你所说的 IDE。IDE 逐渐消失,然后你真的有了这种将代码视为输入的新思维方式。这就是为什么如果我看看新的 IDE,它更多的是关于我如何真正输入规格和 markdown 以及软件工厂周围的一切。所以这种转变改变了。有趣的是,我们正在谈论它。就像我的一个想法是,有了这些决策引擎模型,我终于又回到了 IDE,因为你知道,有趣的是,它有点像函数,对吧?它是嵌入在软件中的智能函数。所以我认为甚至 IDE 也可能重生,因为我们考虑以更 token 高效的方式嵌入智能。所以我不说这是一种永久性的转变。这是关于现在有多种产品形态实际上帮助你提升模型能力。

Yeah, no, it's a great thing to unpack, right? I think to your point, being early to it, when we, you know, this was pre GPT-4, right? So when we even started there, and even just the code completions were magical. That was sort of really what GitHub Copilot, you know, if you remember we worked on IntelliSense forever and we made good progress and then we said wow, this thing is working. And quite frankly, there was a lot of skepticism that it would ever work, but it started and that was even my own belief in scaling laws quite frankly was fundamentally reaffirmed, let's say, when I started seeing that. And I thought we did a good job of trying to build that product. One of the keys of building with models is when model jumps happen non-linearly, the form factor has to change quickly, right? That's the, you know, what is AI first product building in that sense is unforgiving because when you have to basically build everything around a given model and then be ready for the next model to have some emergent capabilities and rewire yourself for that, right? Don't be in love with what you built. So that's why even when agentic coding or task handling started becoming something that one could do, right? It's not just code completions, but you can assign small tasks. That's where Cursor really, you know, they took the IDE, they forked, but they did a good job on really the agent mode. And so we then caught up on the agent mode because that was sort of something that we could do on top of even the OpenAI models. But then the thing that hit us was obviously great work that Anthropic did with their model, which then truly made the CLI as a form factor and then you could even just say just complete full handling of, you know, code went from input to output effectively, right? I mean, that I think was the transformation in 6 months or so. We went primarily that to your point about IDEs. IDEs kind of faded away and then you really had this new way of thinking about code as input. And that's why if I look at the new IDE, it's more about how do I really input the specs and the markdowns and everything else around the software factory. And so that transformation changes. Interestingly enough, it's funny that we're talking about it. Like one of the things thoughts I have is with these decision engine models, finally I'm back again to IDEs because, you know, in interesting way, it's kind of like functions, right? It's intelligent functions that are embedded inside software. So I think there could be a rebirth even of an IDE as we think about even embedding intelligence in a more token efficient way. So I don't say it's a permanent shift in one way. It's about having multiple form factors now that actually help you with model capability.

决策引擎:新基础原语 Decision Engines as a New Primitive

Host

我确实想在这里问你。你提到了决策模型以及你对它们的兴奋。我们俩最近在不同的论坛上与 Diogo 交谈过。你的想法在决策模型上,以及在语言模型对于编程任务、编码、以可靠方式构建软件的局限性上,发生了怎样的变化?

I do want to ask you here. You mentioned decision models and how you're excited about them. Both of us recently talked with Diogo in different forums. How has your thinking changed on both decision models, but also on the limits of LM specifically for programmatic tasks for coding, for building software in a reliable way.

Satya

是的,我想那就是关键,对吧?也就是说,看,我认为 LLMs 非常棒,我们可以谈论甚至你如何知道

Yeah, I think that that's kind of it, right? Which is, look, I think the LLMs are fantastic, and we can talk about like even how you know

Host

我们是工程师。让我们直截了当地说,嘴巴是极好的,对吧?

we're engineers. Let's cut to the point where the mouth is fantastic, right?

Satya

对。从你甚至如何思考软件工厂的意义上说?我们可以深入整个事情。但我喜欢决策引擎的地方在于它是一种新的原语,对吧?以及 Diego 如何谈论它,以及我们自己的决策引擎和 Achint 等人如何谈论它。我喜欢这一点,因为你知道,它有点像 Excel 宏,对吧?你知道,电子表格与宏发生的事情可以发生在代码中,智能作为原语嵌入,对吧?所以我不知道,你构建任何系统。就像我想我正在看一个 Xbox 系统,他们构建的用于管理所有社区反馈的系统,对吧?这是一个经典的社区反馈来了,我基于它做出一组决策。我不想真的在它上面生成 LLM 代码。我只希望能够做出决策。所以更好的做法是嵌入这个原语,对吧?它有一个分布。你可以调用。它有点像一个不需要一次又一次生成的东西。它是 token 高效的。所以我认为这是软件工程仍然非常活跃的地方之一。它在变化。我喜欢 Diego 的事情之一是他想让它像数据库一样无聊。事实上,对我来说,关系数据库是我成长过程中最喜欢的开发者链产品,除了编译器和 Lex 和 Yacc 之类的,就是数据库,因为关系数据库是一种可靠的方式来存储、查询你的信息,使各种应用成为可能。我认为智能时代的决策引擎也会做同样的事情。

Right. In the sense of how do you even think about the software factory? And there's an entire thing that we can go into. But what I love about decision engines is it's kind of a new primitive, right? And how Diego talks about it and how our own decision engine and Achint and others are talking about it. And I like that because you know, it's kind of like, you know, Excel macros, right? You know, what happened to spreadsheets with macros can happen with code with intelligence embedded as a primitive, right? And so I don't know, you build any system. Like I think I was looking at an Xbox system that they'd built for managing all of the community feedback, right? It's a classic community feedback comes, I make a set of decisions based on it. I don't want to really generate LLM code on it. I just want to be able to make decisions. So it's better to embed this primitive, right? It has a distribution. You can make a call. It's sort of a It doesn't need to be generated again and again. It's token efficient. And so I think that this is one of those places where software engineering remains very much alive. It's changing. One of the things I like about Diego's thing is he wants to make it as boring as databases. In fact, to me, a relational database I grew up with my like the most favorite product in the developer chain other than the compilers and the Lex and Yacc and what have you was the database because the relational database is a reliable way to be able to store, query your information just made all kinds of applications possible. And I think decision engines in the era of intelligence will also do the same.

决策引擎:智能时代存储过程 Decision Engines as the Stored Procedures of the Intelligence Era

Host

嗯,事实上,数据库是核心,所以我们谈论软件公司的本质是什么。是业务逻辑,对吧?但当你深入时,它通常在数据库中。事实上,很多业务逻辑过去被编码在存储事务中,一些东西。所以我想知道是否

Well, in fact, the databases are the heart and so we talk about what is the essence of a software company. It's the business logic, right? But when you go deeper, it's often in the database. In fact, so much of the business logic used to be coded in stored transactions, some of things. So I wonder if

Satya

是的,这是一个好问题。

Yeah, it's a good one.

Host

就像这是智能时代的存储过程。

It's like this is the stored procedure of the intelligence era.

Satya

是的,所以我想知道是否有一个类比或希望,因为我确实看到我们还没有找到正确的原语。就像我们知道 LLMs 可以生成代码。我们可以使用 LLMs 做不可靠的事情,但我们仍然缺少这类原语,对吧?

Yeah, so I wonder if there's an analogy or a hope because I do see that we haven't found the right primitive. Like we know that LLMs can generate code. We can use LLMs to do unreliable stuff, but we still are missing these kind of primitives, right?

Host

是的。是的。所以我对此非常兴奋,我认为这是如何开始的,你知道,因为另一个挑战是,你知道,让我们暂时回到软件工厂,对吧?如果你看看甚至 Co Work 是如何构建的,或者甚至今天 GitHub Copilot 运行时是如何构建的,你知道,我认为人们已经记录过它。

Yeah. Yeah. So I'm very excited about it and this is I think the beginning of how, you know, because one of the other challenges, you know, so let's go to the software factory for a second, right? If you look at even how Co Work is built or even the GitHub Copilot runtime is built today and you know, I think people have logged about it.

代码生成与验证 Code Generation and Verification

Host

这很吸引人,对吧?你知道,在一个代码生成变得廉价的世界里,变得困难的是验证,对吧?事实上,我觉得,天哪,我希望,你知道,我想加入这些 Marcus Andernovitch、Scott Hanselman 的,你知道,感知项目之类的,因为我觉得我得学习,因为我还没有在一个现在主要以代码为输出的团队中工作过,对吧?但代码作为输出有一个特定的操作顺序。并不是说代码作为输出就意味着每个人都在生成各种随机代码。就像你必须构建验证的东西。你如何把每一个失败都拿回来,然后构建测试用例,并确保你真正验证了测试,对吧?我的意思是,在管理复杂软件项目方面,人们正在积累太多新的部落知识。事实上,这是一场噩梦。我认为即使在他们进行的 Rust 迁移中,他们也不得不回退。所以这就是所有代码,巨大的优势和一些巨大的劣势。所以这就是为什么,正如你所说,拥有像这个决策引擎这样的额外原语,我认为是好的。

And it's fascinating, right? You know, in a world where code generation has become cheap, what has become hard is verification, right? In fact, one of the things that I feel like, oh my god, I wish, you know, I want to join one of these Marcus Andernovitch, Scott Hanselman, you know, perceptorships or whatever they call because I kind of feel like I got to learn because I've not worked in a team that is mostly now code as output, right? But the code as output has a particular order of operations. It's not like code as output means everybody is just generating all sorts of random code. Like the way you got to build the verification stuff. How do you take every failure and then build back the test case and make sure you truly verify the test, right? I mean, there's just so much, I would say, new type of tribal knowledge people are building up in managing complex software projects. In fact, it was a nightmare. I think even in that rust migration they were having, they had to revert back. And so that is sort of all the code, the great advantages and some massive disadvantages. So that's why, to your point, having additional primitives like this decision engine I think are good.

Host

你提到,作为这家了不起的公司的 CEO,你看到所有事情发生,对吧?你监督,你获取信息,如果可以的话,你会想去团队工作。

And you mentioned that as a CEO of this amazing company, you see all of the things happening, right? You oversee, you get information, you would want to go and work on a team if you could.

Satya

哦,绝对。事实上,你知道,我梦想的下一份工作就是那个,对吧?如果我能回去,只是成为一部分。事实上,有趣的一件事是,你知道,我打算给 Scott 和 Mark 发邮件说,嘿,他们在微软内部创建这个叫做感知项目的新项目,做得非常棒。我想他们甚至和卡内基梅隆大学有关系什么的。那个想法是,你可以让早期职业的人进来,因为我非常关心,嘿,你能让一个早期职业的人进来,加入这些团队,走上学习曲线吗?因为我认为学习曲线,对吧?记得人们过去来微软实习,他们会去看,嘿,Cutler 的代码怎么样?我怎么去看 malloc 的实现?所以我认为我们必须真正发现的是它的新形式。所以是的,如果,你知道,我希望有一天我能再次这样做。

Oh, absolutely. In fact, you know, my dream next job would be that, right? If I could go back and just be part. In fact, one of the things it's interesting, you know, I was going to send mail to Scott and Mark and say, hey, they're doing an unbelievable job of creating this new program inside of Microsoft called perceptorship. I think they even have a relationship with Carnegie Mellon or something. And that idea that you can have people who can come in even early in career because I care a lot about, hey, can you have an early in career person come in, join one of these teams, go up the learning curve? Because I think the learning curve, right? Remember people used to come to Microsoft for an internship, they would go see, hey, how is, you know, Cutler's code? How do I go look at the malloc implementation? So what's the new form of that is what I think we have to really discover. And so yes, if we, you know, my hope would be one day I get to do that again.

Host

我现在明白了为什么这么多 CTO、工程副总裁、高级副总裁现在降职到小团队工作,学习如何用这些新工具构建。似乎很多都在变化。

I'm now understanding why so many CTOs, VP of engineering, SVPs do take step downs to work on small teams to learn how to build with these new tools right now. Seems a lot is changing.

Satya

我绝对。我认为是的,这一切的组织背景都在变化,我认为是的,如果我,如果我考虑一下,我认为这非常重要。事实上,你知道,在许多方面发生巨大变化的时期,更加亲力亲为,对吧?顺便说一句,贴近客户,公开构建,在成功产品或成功技术的各个方面都有很多要学习,所以是的,有很多机会。

I absolutely. I think yeah, the organizational sort of context of it all is changing that I think yeah, if I, if I think about it, I think it's very important. In fact, you know, being a lot more hands on at a time of great change on many fronts, right? By the way, being close to customers, building in the open, there's so much to learn on all aspects of what makes successful products or successful technology, that yeah, there's a plenty of opportunity.

软件工厂 Software Factory

Host

我们刚才一直在讨论“软件工厂”这个词。那我们来谈谈它吧。我觉得这个词有点模糊,但它正在流行起来。对你来说,它意味着什么?

We've now been throwing around the term software factory for a little bit. So let's actually talk about it. I feel this term is a bit vague, but it's coming around. What does it mean for you?

Satya

是的,我也还在努力理解它。所以我在匹配当今软件团队中人们所做的事情的模式,对吧?所以你知道,我想我会让你的观众去看 Bass 写的博客。我的意思是,事实上,有两篇不错的博客,对吧?Co-work 团队和 GitHub Copilot 团队,他们都发布了这些博客。一个是关于一个个体开发者

Yeah, so I'm also still kind of coming to grips with it. So I'm pattern matching in what is teams of people in a software team today do, right? So you know, I think I'd point your viewers to the blog that Bass wrote. I mean, in fact, there are two nice blogs, right? The co-work team and the GitHub Copilot team, both of them have these blogs out. One is what one individual developer

Host

是的。

Yep.

Satya

在 GitHub 那边能够做到的,就是维护整个运行时,并且重新,你知道,我想他们也迁移了。我忘了是从 Python 还是什么,我猜是到 Rust。

was able to do that was on the GitHub side, which is maintain an entire runtime and re, you know, I think they also migrated. I forget whether it's from Python, I guess, to Rust.

Host

总是迁移到 Rust。

It's always to Rust.

Satya

到 Rust。但尽管如此,他们那是一个美丽的案例研究,关于单个开发者的生产力以及进行维护等意味着什么。然后有一个团队。我想是 9 个或 12 个或不管多少个工程师。那个团队和软件工厂的总规模有点疯狂。他们也写了一篇不错的博客,但那里有很多我喜欢的东西。我的意思是,这就是,你知道,如果我能直接加入那个团队,真正第一手学习,而不是仅仅阅读博客或与 Bass 和团队交谈。但例如,据我理解,根本的转变是,对吧?一个是代码从输入变成了输出。很好。然后你不得不说,好吧,那意味着规范和意图,对吧?这一直很重要。设计文档很重要,并且要对齐。所以这现在非常重要。

To Rust. But nevertheless, they kind of that's a beautiful case study of what the productivity of a single developer and what does it mean to do that maintenance and so on. Then there is the team. I think it's nine or 12 or whatever number of engineers. It's kind of crazy that that's the total size of that team and the Software Factory. And they have also done a nice blog, but there's so many things there I love. I mean, this is where, you know, if if I could just go join that team and really learn it firsthand versus just reading blogs or talking to Bass and team. But for example, the fundamental transformation as I understand it is, right? One is code is gone from being input to output. Great. Then you kind of have to say, okay, that means specification and intent, right? Which was always important. Design docs were important and so aligned. And so that that's kind of, you know, very important now.

Host

是的。

Yeah.

Satya

我发现的另一个有趣的事情是,我和一个早期职业的人谈过,他进去了,现在进入一个代码库实际上容易多了,因为当你管理一个软件工厂时,一切都有文档。

The other fascinating thing I'm also finding is and I talked to one sort of early in career person who went in, getting into a codebase now is actually much easier because the reason when you're managing a Software Factory, everything is document.

Host

是的。

Yeah.

Satya

然后你可以问它问题。

And so and then you can you can ask it questions.

Host

耐心的导师,入职伙伴。

the patient mentor onboarding buddy.

Satya

所以你可以,你入职一个代码库比你在人们头脑中搜寻部落知识要容易得多,如果你愿意的话,对吧?所以另一件事是我学到了很多关于验证的东西,对吧?这就是你进行智能体式编码的方式,就像你给它这些小任务。确保它真正完成。你如何进行批评,对吧?关于拥有一个单独的模型家族,有很多话要说。事实上,这是我一直担心的另一件事,就像,嘿,如果有一个模型家族,而这些模型家族不会有所有这些对齐问题,你如何使用批评来确保一个模型家族的挑战不会成为你的挑战。所以无论如何,我有点认为有一种全新的构建方式正在出现,这种新的软件工程,它需要软件工程师。这只是一个交互层次。也许这就是为什么我喜欢甚至这个类比,你知道,当你从编写自己的汇编到说哦哇我有一个编译器,突然你觉得哦哇我现在可以从意图到代码,我不需要知道汇编。有一点那样。这是一个抽象层次,但甚至你在团队中工作的方式也改变了,你构建产品的方式也改变了。另一个我真正兴奋的是来自生产的追踪回来。我也喜欢那个,对吧,你知道

And so you can have you your onboarding onto a codebase is so much more easier than you scrounging around for tribal knowledge that's in people's heads, if you will, right? So the other thing is I'm learning a lot about this verification stuff, right? Which is the way you want to do agentic coding is like these little tasks that you give it. Make sure that it's truly done. How do you have the critique, right? There's so much to be said about having a separate model family. In fact, that's the other thing that I I've always worried about is like, hey, if there's one model family and these model families are not going to have all these alignment issues, how do you use a critique to ensure that one model family's challenge doesn't become your challenge. And so anyway, I I kind of look at saying there's an entire new way to build that is emerging and this new software engineering and it requires software engineers. It's just a level of interaction. Maybe that's why I liked the even the analogy of you know when you know you went from writing your own assembly to saying oh wow I have a compiler and suddenly you felt like oh wow I can now go from intent to code and I don't need to know assembly. There's a little bit of that. It's a level of abstraction but even how you work in a team has changed and how you construct a product has changed. The other one that I'm really excited about is the the traces from production coming back. I love that too right which is you know

Host

可观测性,但现在我们有更多更多的机会

the observability but now we have more more opportunities

Satya

是的,我的意思是当人们说 RSI 时,那有点像是我的 RSI,对吧,就是能够获取我的生产日志,然后能够验证修复。

Yeah I mean when people say like you know RSI that's kind of my RSI right which is it's sort of the ability to take my production logs and then to be able to verify the fix.

可验证循环与智能体运维 Verifiability loop and agentic ops

Satya

事实上,那个可验证性循环很棒——你可以部署修复并一路部署下去,而且它确实通过了所有测试。这真的非常有用。所以我认为这些团队有很多经验教训,并在微软内部传播。我的意思是,看到这现在成为团队构建的方式,真是太好了。

In fact, that verifiability loop is great—you can then deploy the fix and deploy it all the way through, and it actually passed all the tests. That actually is very helpful. And so I think there's a lot of learning that these teams are having and spreading across Microsoft. I mean, it's just good to see how this is now becoming the way teams are structuring.

Satya

另一个我喜欢的是我们的网络——Azure 网络团队。实际上,这曾经是我最喜欢的故事——这个团队说,嘿,我们——你知道,微软为我们的全球网络管理着大约 500 家光纤运营商。你知道,总是有物理资产——总有什么地方出问题。你知道,管理这一切的时髦词是 DevOps,但现实是电子邮件和各种临时流程。所以他们构建了整个智能体系统,实际上执行运维,而且运维甚至连接到他们的代码库。所以现在不仅仅是代码库本身或代码基础,甚至还包括围绕它的运维。所以你可以开始看到,技术团队工作的本质是构建这些运维加软件的工厂来完成任务,而不是直接完成任务。但他们仍然在做一项略有不同的工作。

The other one that I love is our networking—Azure networking team. Right, this—in fact, this used to be my favorite story—where this team said, hey, we're—you know, Microsoft manages something like 500 fiber operators all over the world for our global network. And you know, always it's a physical asset—something or the other goes wrong. You know, the fancy word for managing all that is DevOps, but the reality is it's emails and you know all sorts of ad hoc processes. And so they went and built this entire agentic system that actually does the ops, and that ops is connected even to their repos. And so now it's not just the repo on its own or the code base, but it's even the operations that around it. And so you can start seeing how the nature of what technical teams do is build these factories of ops plus software that do the job versus doing the job. But still, they're doing a job which is slightly different.

平衡实验与产品团队 Balancing experimentation and product teams

Host

我确实想知道——因为在 AI 实验室里这很常见,而且他们非常开放地分享那里的软件工程师的工作如何变成让智能体和循环做得更好。所以就像你刚才说的。但仅从实际层面来说,像在微软这样的地方,你会有可以实验的团队,他们正在构建新工厂,或者是平台团队,也许他们有额外的周期,他们在做所有这些事情。但你也有产品团队,他们实际上有产品,有真实客户。这就是我看到很多公司挣扎的地方——他们确实有真实的业务,他们看到 AI 实验室的这些理想化的事情,或者甚至他们现在正在听你说话,他们在想,等等,我不能就这么放下一切去建一个工厂,因为我拿这个东西怎么办?你如何思考这种平衡,尤其是在微软内部?你告诉工程师什么?比如,你的工作如何变化,尤其是当你在做一个好产品、一个盈利的产品,客户依赖它的时候?

I do wonder—because in AI labs it's pretty common, and they're very open about sharing how software engineers there, their job is becoming to allow the agent and the loop to do a better job. So, so what you just said. But just sticking to practical level, like at a place like Microsoft, you will have teams that can experiment and they're building new factories or their platform teams and maybe they have extra cycles and they're doing all these things. But you also have product teams who actually have a product, you have real customers. And this is where I see a lot of companies struggle—who do have a real business and they're seeing these aspirational things from AI labs or maybe even they're listening to you right now and they're thinking, hang on, like I can't just drop everything and build a factory because what am I going to do with the thing? How are you thinking about both this balance and especially inside of Microsoft? What do you tell engineers? Like, how is your job changing, especially when you're working on a product that is a good product, is a profitable product, customers depend on it?

客户价值与避免随机AI Value to customers and avoiding random AI

Satya

是的,这是一个很好的观点。我的意思是,首先,你知道,我们不会得到许可去做那些不为客户增加价值的事情。让我们从这里开始。关于你提到的一点——我们也犯过不少错误,你知道,去做随机的 AI 事情,只是因为它们很酷。你最终会——

Yeah, it's a great point. I mean, like first of all, you know, we don't get permission to do things that are not adding value to customers. Like let's start there. To your point about a little bit of it—and we made our share of mistakes of you know, going and doing random AI things just because they were you know, you know, cool to do. You kind of ultimately are going to—

Host

我想每个人都做过。

I think everyone's done that.

Satya

每个人都做过,我也做过不少,希望从中学习并变得更好。但话虽如此,如果你说好吧,让我们甚至——让我们面对现实,对吧?今天当我查看 Windows 中的 CVE 时,对吧?因为你现在有 AI 模型在发现关键漏洞。是的。所以我们现在需要修复所有这些问题,因此有了新的管道。所以事实上,如果我看看甚至 Windows 从崩溃跟踪到修复,他们基本上在构建一种新型的软件工厂。只不过这是一个棕地,对吧,Windows 的构建系统仍然是 Windows 的构建系统。事实上,如果你想想所有这些棕地项目,最困难的部分是构建和 CICD。然后当然,你必须真正,你知道,甚至,你知道,如果我看看甚至 GitHub 下面的变化量,对吧,如果你想想,对吧,就像,你知道,我们有过,我的意思是,突然所有这些编码智能体出现在 GitHub 的工作中,规模已经飙升。

Everyone's done that and I've done my fair share of it and learned from it hopefully and will get better. But that said, if you sort of say well, let's take even—let's face it, right? Which is today when I look at the CVEs in Windows, right? Because you now have AI models that are finding the critical vulnerabilities. Yep. And so we now need to go fix them all and so therefore there's a new pipeline. So in fact, if I look at even the Windows from crash traces all the way to fixes, they are building out essentially a new type of software factory. Except this is a brownfield, right, which is the build system of Windows is still the build system of Windows. In fact, if you think about in all these brownfield projects, the most difficult part is builds and CICD. And then of course, you got to really, you know, even, you know, if I look at the amount of change that underneath even GitHub, right, which is if you think about it, right, like, you know, we've had, I mean, suddenly all these coding agents showed up at work at GitHub and the scale has gone through the roof.

超人规模与重思原语 Superhuman scale and rethinking primitives

Host

超人类水平。

The superhuman level.

Satya

是的。因此,我们必须从根本上改变整个 Git 协议、存储系统。所以我们正在努力解决每一个部分,以确保我们生活在棕地中,而不是——我的意思是,绿地的东西我们总是会做。但现实世界全是棕地,棕地需要逐步地,你知道,系统的加固,重新思考核心原语,对吧,从源代码控制到构建再到 CICD,这些在棕地项目中都有效。这需要时间。需要投资。需要思考,操作顺序或操作次序非常非常关键。

Yeah. And so therefore, we have to fundamentally change the entire the Git protocol, the storage system. So we are working on every part of this to make sure that we are living in the brownfield as opposed to—I mean, the greenfield stuff we will always do. But the real world is all about brownfield and the brownfield requires stepwise, you know, hardening of systems, rethinking of the core primitives, right, all the way from source code control to build to CICD that works in a brownfield project. And that takes time. That takes investment. That takes thought and the sequence of operations or the order of operations is very, very critical.

全员普及新工具链 Democratizing new toolchain across company

Satya

所以我希望我们团队做的一件事——不是关于——我喜欢这次转型的一点是,这不是关于,哦,这里有一些新的闪亮的东西,这里是做那个的团队,其他人生活在旧世界里。就像,公司里的每个人都可以使用新的工具链和新的工作方式,并应对他们面前的任何产品的挑战,无论是 Windows 还是 SQL Server 或云原生服务等等。我认为这就是我们正在看到的。

So one of the things that I want our teams—is is not about the one nice thing I like about the transformation is this is not about, oh, here is some new shiny object and here is the team that's doing that and everybody else lives in the old world. Like, everybody across the company gets to use the new tool chain and the new ways of working and tackle the challenge in front of them on any product, whether it's Windows or SQL Server or a cloud native service, what have you. And that I think is what we're seeing.

与过往技术采纳对比 Comparison to past tech adoption

Host

公平地说,这听起来和我们过去做的并没有那么不同,因为当有新技术时,比如说 React,这是一种新的前端技术,每个人都想用它。我们所做的是,没有人被允许停下来迁移到 React。但当你构建新产品或添加新功能时,你开始加入它,你开始混合它。所以也许这和我们采用它的方式并没有那么不同,但这有什么不同呢?是品味吗?

To be fair, like, this doesn't sound all that different to what we did in the past because when there was a new technology, let's say React, which was a new front-end technology, everyone wanted to use it. And what we did is no one got permission to just stop and migrate to React. But when you built the new product or you added a new feature, you start to add it in, you start to mix it. So maybe it's not all that different to how we're adopting it, but what what is different about this? Is the taste?

运算顺序与抽象的物理 Physics of order of operations and abstraction

Satya

说得好。就像,你知道,有时我,你知道,我有点说,是的,我以前见过这个等等。但我认为这次我确实感觉到不同的一件事是我们做事方式的物理规律,对吧?那个操作顺序,抽象层次。也许事实上,我认为一个迷人的事情是就像人们移动时,对吧?我加入劳动力大军时正处于我们从,你知道,事实上,在大学里我主要使用 VAX,对吧?你知道,用 VT VT,你知道,终端 uh VT

Good one. Like, you know, sometimes I I, you know, I kind of say, yeah, I've seen this before and what have you. But I think the one thing that I do feel that's different this time around is that the the physics of how we do things, right? That order of operations, the level of abstraction. Maybe In fact, one of the fascinating things I think about is it is like when people moved, right? I I I joined the workforce at the cusp of when we went from, you know, in in fact, in college I was mostly working on a VAX, right? You know, with a VT VT, you know, terminal uh VT

Host

这是大型机。

This is the mainframe.

Satya

呃是的,或者小型机,对吧?VMS 系统,对吧?所以然后然后我实际上在 Sun 的工作站上工作了几年,然后我来到微软,所以然后,你知道,世界有点稳定了,对吧?你主要在 IDE 中写代码。在我的职业生涯中,它并没有真正改变。我的意思是,新技术出现了,但所有那些新技术都是新框架和新运行时。这不像概念数量发生了根本变化。甚至移动设备也像是小型化的 PC,带有新的用户界面,对吧?和新的框架。

Uh yeah, or a mini, right? VMS systems, right? So and then and then I joined actually for a couple of years at Sun on a workstation and then I came to Microsoft and so and then, you know, the world kind of was stable, right? You mostly wrote code in an IDE. Throughout my professional career, it sort of didn't really change. I mean, new technologies came, but all those new technologies were new frameworks and new runtimes. It was not like the concept count radically changed. Even the mobile thing was sort of a miniaturized PC with new sort of uh user, you know, user interface, right? And new frameworks.

软件工程格局之变 The changing software engineering landscape

Host

这是一个所有东西都要写清楚的地方——运行时是什么?框架是什么?IDE 是什么?技术栈是什么?——把每一部分都写出来。

This is a place where everything—what is the runtime? What is the framework? What is the IDE? What's the stack?—write every single part of it.

Satya

我觉得,更多的是你思考方式上的一点变化。然后别忘了,新模型可能随时出现,对吧?无论是新的前沿模型,还是像决策引擎这样的新东西,它可能再次改变,对吧?这就是我认为我们也必须准备好,整个范式随时可能被彻底改变。所以,至少我会思考这两件事:回顾过去,我们个人、公司或团队在管理变革时,什么时候处于最佳状态,这总是很重要的;但同时,保持开放心态,认识到这次可能不同,也是好的。

A little more of a change in how you think about it, I think. Then let's not forget a new model could drop, right? Whether it is a new frontier model or a new thing like decision engine, and it could change again, right? This is where I think we also have to be ready for the regime to completely be transformed at any moment, if you will. And so therefore, I think these are two things at least I think about—thinking of it, learning from the past, when were we at our best individually or as companies or as teams in managing change, I think is always important, but also being open-minded that this time is different is also good.

Host

是的,而且特别是对我们软件工程师来说,最大的变化是——你知道微软可能大部分员工都是软件工程师或接近代码的人——我们过去大部分时间都在写代码,大部分努力都是在大学和之前的工作中学习,以擅长于此。而这一部分突然消失了,我知道对很多人,包括我,早些时候很难接受,因为我们从未见过任何技术变革会这样——它总是增加一个新工具。你如何看待微软内部或外部的人,那些真正优秀的工程师,他们花了几十年变得非常擅长编程,这很难做到,对吧?他们如何接受这一点?

Yeah, and also specifically for the biggest change for us software engineers—and you know Microsoft has probably majority of the workforce is software engineers or people close to the code—is we used to spend most of our time writing code and most of our effort was learning at college at previous jobs to be good at this. And this part suddenly disappeared, which I know for a lot of people, including me, earlier it was hard to reconcile because we've never seen it with any technology change—it always added a new tool. How do you see people inside of Microsoft or outside who talk with reconcile this, who are really good engineers and they spend decades getting very good at coding, which is hard to do, right?

Satya

是的,我认为没错。所以这就是我至少观察到的,而且当我与那些曾经是优秀软件工程师、现在在新生产模式下也高产的人交谈时,发现仍然需要同样的技能。只是你思考的顺序、工具链变了。这也是另一个原因,我认为这不是贬低专业知识,而是真正利用专业知识,确保你能用新的杠杆来发挥它。这就是工作性质必须改变的地方,而不是认为软件工程不再重要。我认为我们正处于试图弄清楚如何评估新事物的阶段,而在评估新事物的过程中,你突然也会开始评估旧事物,如果这说得通的话。当我观察人们时,显然有 AI 原生的人出现——事实上很有趣,上次 GitHub Universe 我去了,遇到了一群大学生,那是第一批在 AI 时代成长起来的大学毕业生,对吧?因为他们上大学时就有了 GitHub Copilot——所以他们不知道没有 GitHub Copilot 的 IDE 世界。这让我震惊,事情就是这样。所以会有一些人天生就熟悉这个工具链。但我也认为,当我在微软看到,任何在某个领域有深厚专业知识、有好奇心、正在学习新工作方式的人,都绝对在其中蓬勃发展。所以我希望确保的一件事是,我们为人们创造这样的机会,因为归根结底,重要的是你职业生涯的经济回报。因此,我们希望确保在微软,你既能重视你拥有的技能,又能建立市场将重视的新技能。

Yeah, no, I think that's right. So this is where what at least I am observing, and as I talk to even people who have been great software engineers previously and now are prolific even in the new mode of production, is the same skills are still required. It's just again that what you think about—the order of operations, the tool chain—has changed. And so that's one of the other reasons where I don't think it's about devaluing the expertise; it's about really taking the expertise and making sure that you can exercise that with new leverage. That is where the nature of how you work has to change, as opposed to thinking that this is about software engineering no longer being important. And I think that we are in that phase where we are trying to figure out how to value the new thing, and in valuing the new thing you are suddenly going to even start valuing the old thing, if it makes sense. When I look at the people, there's obviously somebody who's AI native who comes out of—in fact, it is funny, last GitHub Universe I went down and I met a bunch of college kids, and that was the first batch of college graduates who grew up in the AI era, right? Because they went to college when GitHub Copilot—so they didn't know an IDE world pre-GitHub Copilot. It was stunning to me that that was sort of what's happening. So there is going to be people who are going to come at this who are going to be native to this tool chain. But I also think that people—when I see at Microsoft, anyone who has great depth and expertise in any part has curiosity and is learning the new ways to work—are absolutely flourishing in it. And so one of the things I'm hoping to ensure is we create that opportunity for folks because at the end of the day, it's about economic return on your professional careers that matters. And so therefore we want to make sure at Microsoft you can both value the skill set you have and build the new skill set that the marketplace will value.

自愿退休与AI效率 Voluntary retirement and AI efficiency

Host

我真的很喜欢你对此的看法,也喜欢你谈到初级新毕业生、早期职业人士如何做出贡献、如何快速上手。但有一件事在深度圈子里流传,我就直接向你提出来:人们在猜测,为什么微软要提供自愿退休方案?这意味着什么?有一种恐惧是,这可能是由于 AI 使团队更高效而潜在裁员的先兆?我就想直接问你这个问题。微软从未有过这样的事情,这实际上是我最关注的,还有它是自愿的。你的想法和公司的想法是如何在这个问题上达成一致的?

I'm really liking how you're looking at this and also how you talked about how junior new grads, early career people are contributing, they can get up to speed. But one thing that's been going around in in-depth circles and I'll just lay it out for you is people are speculating, why is it that Microsoft is making the voluntary retirement offers? Like what does this mean? There's this fear that could this be a prelude to potentially a layoff because of AI making teams more efficient? And I just want to ask you that on the head. There's never been anything like this at Microsoft, which is what I actually paid most attention to and also the fact that it is voluntary. How did your thinking and the company's thinking come together on this one?

Satya

这是一个非常重要的问题。我们讨论了很多,我们觉得让我们承认两件事,对吧?一是——或者也许是三件事。一是让我们承认,未来会有很多机会。机会的形状可能不同于现在的机会,对吧?如果我们甚至狭隘地只考虑我们刚才讨论的软件工程。

It's a very important question. And one of the things that we discussed a lot and we felt was let's acknowledge two things, right? One is that—or maybe three things. One is let's acknowledge that there is going to be plenty of opportunity going forward. The shape of that opportunity may be different than what is the opportunity, right? If you let's even narrowly think about just software engineering as we were discussing.

Host

是的。

Yeah.

Satya

所以我们就从那里开始。

So let's start there.

Host

是的。

Yeah.

Satya

然后我们应该说的下一件事是,好的,很好。让我们在公司内部创造更多这样的新机会,对吧?这就是我们谈论的一切,无论你是在做 Windows 还是做 AI 训练,坦率地说,你都可以真正装备自己新的技能,这些技能在微软内部和市场上都将是有价值的,对吧?

Then the next thing we should say is okay, great. Let's create more of that new opportunity inside the company, right? That's where everything that we're talking about, whether you're working on Windows or whether you're working on AI training, quite frankly, you get to really equip yourself with new skills that are going to be valuable inside of Microsoft and in the marketplace, right?

Host

是的。

Yes.

Satya

但这个新事物的形状将不同于以前的形状,对吧?团队规模、团队工作流程。所以我们需要采取——并不是我们不想让微软的 20 万人蓬勃发展,所有人都成长。但微软的 20 万人及成长将必须以不同的方式配置,对吧?当我观察微软内部时,甚至一些经济回报,对吧?如果你看系统工程师的工资,对吧?或者基础设施人员,那是非常高的。如果你是一个训练有素的研究员,也是一样,对吧?所以一些专业知识竞争非常激烈,价值很高,对吧?因此,坦率地说,我们想给予更多的自主权。事实上,这个自愿方案主要是关于让每个人都有更多的自主权,把自己安排到对他们长期经济机会有意义的地方。我认为这是关键。你知道,我当 CEO 很久了。在那之前我是高级主管等等。但归根结底,我从一个个人贡献者成长到这个位置,对吧?所以我经常思考当我还是个人贡献者或领导,或者领导的管理者时,那是我思考如何在公司内部创造机会的视角。我不想被困在一个不转型的团队里。

But the shape of what this new thing looks like is going to be different than what the shape of what it was before, right? Team sizes, team workflows. So we need to take—it's not that we don't want the 200,000 people at Microsoft flourishing, all of them, and growing even. But the 200,000 people at Microsoft and growing will have to be configured in different ways, right? When I look at even inside of Microsoft, some of even let's say even the economic return, right? If you look at the wages for a systems engineer, right? Or an infrastructure person, it's through the roof. Same thing if you are a researcher in training, right? So some of the expertise is highly competitive, highly valued, right? And so we want to give more agency, quite frankly. In fact, this voluntary thing was mostly about let's give everybody more agency to get themselves slaughtered into places that make sense for their economic opportunity long term. That's I think the key thing. You know, I've been CEO for a long time. I was a senior exec before that and what have you. But at the end of the day though, I did grow up as from an IC to this, right? So I think about a lot of the times when I was an IC or a lead, right? And maybe a manager of leads, that's sort of the lens through which I think about how should we create opportunity inside the company. I don't want to be stuck in a team that's not transforming.

微软内部的变革与能动性 Change and agency inside Microsoft

Satya

即便在微软内部,我们得承认,我没法坐在这里说微软的每个部门都真的领会了精神、都在按我希望的方式转型。如果你所在的团队让你觉得职业上卡住了,你可以用脚投票。我不希望这样,因为坦白说,微软内部本身就有很多地方可去。所以这也是我们想创造的另一种机会。在这个背景下,我们正努力给人们更多自主权,因为这是一个动荡的时期。我希望能坐在这里说一切都会照旧,但我不能。这正是我想对变化保持透明的地方。我想确保我们踏踏实实做好在公司内部创造各种机会的苦活,然后确保我们的员工能真正在这里实现职业成长。

Even inside of Microsoft, let's face it. It's not like I can sit here and say every part of Microsoft has really got the memo and they're all transforming the way I want them to. If you're in a team where you feel stuck professionally, you can vote with your feet. I don't want you to, because quite frankly, there are many places you can go inside of Microsoft itself. So that is another opportunity we want to create. In that context, we are trying to give more agency to people, realizing that this is a time of tumult. I wish I could just sit here and say it'll all be the same. I cannot. This is one of those places where I want to be as transparent about the change. I want to make sure we do the hard work of creating all the opportunity at the company, and then make sure that our workforce is able to really make their professional growth come true here.

Host

我觉得你说得很客气、很周到,但换成我的话:世界已经大变样了。比如我们看到 Claude Code 带来了如此迅速的变化,那些在方向盘上睡着的团队,你是承受不起的。所以你要么跟上,意识到世界变了,我作为软件工程师的工作会变、我的工具会变,要么我接受并想办法对此充满热情,看到所有机会、看到搭建这些工厂和新工作流带来的所有工程挑战;要么我决定,我不干这个。大多数地方你基本是卡住的,但你实际上给了人们一个选择:如果你想从这儿退出,你现在就可以,而且不会有任何芥蒂。

I mean, I think you're saying this very nicely and thoughtfully, but how I would translate it is: the world has changed big time. We saw, for example, Claude Code has been such a fast change, and the teams who fall asleep at the wheel, if you will, you cannot afford it. So you need to either go with this, realize that the world has changed, my work as a software engineer will change, my tools will change, and either I'm on board with this, I figure out how to be enthusiastic about this and see all the opportunities, all the engineering challenges with building the factories, with the new workflows, or I decide, well, I'm not going to do that. And most places you're kind of stuck, but you're actually giving an option for people to decide: if you would like to check out from there, you can, right now, and there's no hard feelings.

Satya

我想是的。我也觉得,说到底,我很在意这一点:我们让人们产生了巨大的焦虑,巨大的焦虑,让他们觉得自己无法掌控自己的命运。

I think so. And I also feel like, look, if you come down to it, and I care a lot about this, we have created enormous anxiety in people, enormous anxiety, sort of feeling that they don't have control of their own destiny.

Host

嗯。

Mm.

Satya

因为我们谈论这场 AI 革命的方式是,当人们说,嘿,也许奇点已经来了,因此也许我们正在建造的会是只有一家公司,或者会没有工作。所有这些都会让哪怕想改变的人也感到焦虑。

Because the way we have talked about this AI revolution is, when people say, hey, maybe the singularity is here and therefore maybe we are building, there'll be one company or there'll be no jobs. All that sort of leads to anyone who even wants to change being sort of anxious.

Host

是的。压力。

Yes. Stress.

Satya

所以我觉得,这正是……你看,我们是一家 20 万人的公司。我知道我们每个人做的事情都会发生很大变化。想想看,微软增长最快的岗位之一是 FDE。

And so I kind of feel that this is where, look, we're a company of 200,000 people. I know that there's going to be a lot of change in what each of us do. Like think about it. One of the fastest growing jobs in Microsoft is FDE.

Host

微软也是。是啊,现在整个行业都是。

Microsoft as well. Yeah, across the industry right now.

Satya

整个行业,因为某种意义上你想想,你有客户代表,非常重要。客户代表旁边是我们的客户,也就是技术支持人员。然后是客户成功,我们称之为客户成功。再然后是技术支持人员,然后还有 FDE 团队。所有这些都需要一个完整的——它们都应该接受和任何软件工程师一样的学徒训练。因为你们都可以——我知道连 LinkedIn 内部也做了这个改变,他们称之为全栈构建者,把角色合并了。所以所有这些变化,我认为都有机会。但你必须走那条艰难的路,真正做结构性调整,把人安排到位,给他们机会,给他们薪酬支持,知道如何评估人们在新岗位上的生产力和卓越表现。这是任何在位者都要做的苦活——所以这就是为什么对新事物的庆祝应该一直在,竞争性市场就是这样运作的。你看,我们是一家 51 年的公司。我们要对自己的样子感到自在。我们要好好服务客户。我们要为这家公司里做出卓越工作的员工创造极好的经济机会,因为我们赋能个人和组织的使命。

The industry, because in some sense if you think about it, you have an account rep, very important. Then right next to the account rep is our customers, the technical support folks. And the customer success is what we call customer success. And then we have even the technical support people, and then you have the FDE team. All of these things require a complete—they all should be going to the same apprenticeship that any software engineer is going through, if you will. Because you can all—I know even in LinkedIn I think they made this change inside where they called them full stack builders, which they combine roles. So all that change, I think there's opportunity. But you have to walk the hard walk of really making the structural changes, getting people slotted, giving them opportunity, giving them the wage support, knowing how to evaluate people with their productivity and their excellence in their new jobs, in their new slots. This is the hard work any incumbent—so that's why the celebration of the new, which always should be there, that's how competitive marketplaces work. Look, we are a 51-year-old company. We're going to be comfortable in our skin. We want to serve our customers well. We want to create fantastic economic opportunity for the employees at this company doing great work, given our mission to empower people and organizations.

Host

另一方面,我只想把这一点记录下来。我们谈到招聘应届毕业生,所以你仍然在招应届生、初级员工,你在他们身上投资。你看到了这一点,因为很多其他公司和初创公司的一个担忧是,他们放缓甚至停止了招聘,因为他们想,有了 AI,最能高效使用它的人,正如我们之前提到的,是那些有专业知识并能加以利用的人,所以他们非常犹豫招任何刚入职场的人。

On the other side of this, but I just want to get this on the record. So when we talked about hiring new grads, so you are still hiring new grads, junior people, you're investing in them. You're seeing that, because one of the worries in a lot of other companies and startups is they've slowed down or even stopped hiring because they're thinking, well, with AI, the people who can use it the most efficiently, as we previously mentioned, are the ones who have expertise and they can leverage that, and they're very hesitant to hire any new career people.

Satya

说得好。我们正在尝试的一件事是——在这个变化中,这就是为什么我认为有些趋势要稍微纵向地、而不只是一年地仔细看,因为我们正处于大量变化的过程中。所以第一件事是确保内部所有人都被重组到合理的位置,这只是第一步。我的意思是,即便在一个团队内部,你也要先把配置弄对。然后才有意义去考虑——哦,行业招聘或新招的大学毕业生,你加进来。所以正在发生的一部分情况,我认为是放缓,也许是放缓来把那些所谓的软件工程新工厂建起来,让新的输入——不只是 token,还有人——能够进来,然后在这个新配置中给你高杠杆。我的意思是,面对现实吧,什么是 AI 原生公司?他们从一份 token 预算和一个人开始,对吧?然后他们加上那个人,用满他的 token 预算,然后说再加下一个人,然后那个人得到一份 token 预算,然后他们的工厂得到预算。

Great point. So one of the things that we're trying to do is—in this change, this is why I think a little bit of the trends have to be watched carefully over slightly longitudinally versus just one year, because we're in the process of a lot of change. So first job is to make sure internally all of the folks are getting re-orged to places that make sense, like just step one. I mean, even inside of a team, you just want to get the configuration right. Then it makes sense to—oh, what is the industry hire or a new college hire you add to this. And so a little bit of what's happening, I think, is the slowdown, maybe a slowdown to get those quote-unquote new factories of software engineering built, such that new inputs—not just tokens, humans—can come in and then in this new configuration give you a high leverage. I mean, let's face it, what is an AI-native company? They start with a token budget and a human, right? Then they add that person, maxes their token budget, and then says let's add the next human, and then that human gets a token budget, and then their factory gets the budget.

Host

是的。

Yes.

Satya

这就是建构主义的方式,并不是说——好吧,那些公司如果成功,有一天也会变大,他们会的,每家公司都会开始看起来像那样,每个项目团队或产品团队都会像那样。所以那个重新配置、新发现和配置的过程需要时间,这就是为什么我认为试图画太多线当作长期趋势,为时过早。

That is the constructionist way, and it's not like—okay, those companies, if they succeed, they will also be big one day, and they will, and every company is going to start looking like that, and every project team or product team will look like that. And so that process of reconfiguring, new discovery and configuration, it'll take time, and that's where I think trying to draw too many lines as long-term trends, I think it's too early.

Host

在我们回到对话之前,我想谈谈 Linear,我们这一季的赞助商。直到最近,我还把 Linear 看作一个帮助规划和组织工作的工具,但它远不止于此。很多团队用 Linear 把变更从想法一路推进到在 GitHub 合并,全程不离开这个工具。我和 Linear 团队的一位工程师 Dima 坐下来聊了聊,看看 Linear 内部的开发者如何用 Linear 构建整个 Linear。所以这里有个真实的问题。

Before we get back to the conversation, I'd like to talk about Linear, our season sponsor. Until recently, I was thinking of Linear as a tool that helps plan and organize the work, but it's much more than that. A lot of teams use Linear to ship changes from idea to merge in GitHub, all without leaving the tool. I sat down with Dima, an engineer on the Linear team, to see how devs inside of Linear use Linear to build all of Linear. So here's a real issue.

赞助商:Linear Sponsor: Linear

Host

当你提交评审时还留着未发送的草稿评论,Linear 会提醒你这些评论的存在,但它不会告诉你这些评论在哪里。有位客户抱怨了这一点,Dima 决定修复它。但为了做这个修复,Dima 并没有打开他的 IDE。相反,他打开 Linear,把问题描述给智能体。智能体构建了修复,创建了 PR,并附上 Storybook 截图以便轻松验证。你可以看到修复效果,草稿评论就链接在截图上。在 PR 里,你可以从 Linear 一键打开预览,亲手试试这个改动,确认它确实生效了。接下来是我最喜欢的部分:引导式评审,目前还在开发中。引导式评审是一种全新的代码评审体验,重要的东西会自己跳出来。评审顶部有摘要和改动示意图。而 PR 被分成若干章节,带你逐步了解改了什么。这些区块以更易理解和评审的方式突出重要改动。哦对了,当 Dima 完成这一切后,他直接自我批准代码,就合并进了 GitHub,全程没有离开 Linear。如果只记住一件事,那就是:Linear 不再只是规划工单、组织工作的地方。它是所有工作发生的地方:创建 issue、启动智能体、查看预览、做评审、合并到生产。Linear 是为团队和 AI 智能体打造的产品开发系统,我很喜欢这个演进。去看看,并在 linear.app/pragmatic 获取六个月试用。

When you submit a review with unsent draft comments, Linear warns you about these comments, but it didn't tell you where those comments were. A customer complained and Dima decided to fix it. But for the fix, Dima did not open his IDE. Instead, he opened up Linear and described the issue to the agent. The agent built a fix, created a PR, and attached Storybook screenshots for easy verification. You can see the fix with the draft comments linked on the screenshot. In the PR, you can open the preview with one click from Linear to play around with the change and see that it worked. And here's my favorite part, guided reviews, which is work in progress. Guided reviews is a new code review experience where the important stuff just pops out. The top of the review has a summary and a diagram of the change. And the pull request is divided into chapters that walk you through what has changed. These blocks highlight the important changes in a way that's easier to understand and review. Oh, and when Dima is finished with all of this, he just self-approves the code and it's merged into GitHub, all without leaving Linear. If you take one thing from this, Linear is no longer just a place for planning tickets and organizing the work. It's where all the work happens: creating the issue, kicking off the agent, checking out the preview, doing the review, and merging to production. Linear is the product development system for teams and AI agents, and I'm loving this evolution. Check it out and get a six-month trial at linear.app/pragmatic.

赞助商:WorkOS Sponsor: WorkOS

Host

如果你做过 B2B 软件,你就知道这套流程。客户喜欢这个产品,然后他们的安全团队发来一张清单:SSO、SCIM 预配、基于角色的访问控制。这些都不是你最初想构建的东西,但交易就卡在这上面。WorkOS 是覆盖所有这些的认证基础设施,这样你团队里没人需要去调试客户的 SAML 配置。这个问题的新版本是智能体。一旦智能体在客户的系统里行动,企业就想知道它被允许做什么。这就是 WorkOS AirLock 的用途。它是基于意图的访问控制。大多数访问控制只问你是谁。AirLock 还会问智能体在试图做某事的当下想做什么。像“智能体不得删除云资源”这样明确的规则以普通代码运行。对于更模糊的判断,由 AI 裁判在运行时评估策略。你用自然语言写这些策略,Claude 可以把你在 Notion 和 Jira 里已有的策略拉出来并导入。你的智能体,无论是 Claude 还是 Devin,都连接到一个单一访问点,你的 MCP 服务器位于其后。AirLock 是你现有权限之上的新一层,所以智能体永远无法获得比调用它的人更多的访问权限。当出现风险时,AirLock 可以升级给人类审批,比如通过 Slack。理念是规则写一次,人类注意力只花在真正需要的判断上。而且每个请求和决策都会被记录。访问 workos.com/airlock 了解更多。

If you've ever worked on B2B software, you know how this goes. A customer likes the product, then their security team sends a list: SSO, SCIM provisioning, role-based access. None of it is what you set out to build, but the deal waits on it. WorkOS is the auth infrastructure that covers all of that, so nobody on your team has to debug a customer SAML setup. The newer version of the problem is agents. Once an agent is acting inside of a customer's systems, the enterprise wants to know what it's allowed to do. That's what WorkOS AirLock is for. It's intent-based access control. Most access control only asks who you are. AirLock also asks what the agent is trying to do at the moment it tries to do it. Clear-cut rules like no agent deletes cloud resources run as plain code. For the fuzzier calls, an AI judge evaluates the policy at runtime. You write these policies in plain English and Claude can pull the ones you already had out of Notion and Jira and import them for you. Your agents, whether that's Claude or Devin, connect to a single access point and your MCP servers sit behind it. AirLock is a new layer on top of your existing permissions, so an agent can never get more access than the person who invoked it has. When something risky comes up, AirLock can escalate to a human for approval, say over Slack. The idea is to write the rules once and spend human attention only on the calls that need it. And every request and decision is logged. Visit workos.com/airlock to learn more.

开放模型趋势 The open model trend

Host

现在我们知道,每当有人在构建他们的软件工厂时,他们都在改变工作方式。现在有一个大趋势,我们稍微深入一点看这是怎么做的——我在整个行业看到的一个大趋势,我相信你也看到了:很多科技公司也在转向开放模型。他们中很多人自己在搭建推理,或者像 Replicate 这样的公司在购买推理,他们买 B200,然后直接在上面部署模型。他们各有各的理由,但从你的视角看,比如微软也在通过 Foundry 做开放模型服务。这个趋势有多大,它是否不可避免?我在问自己,开源运动和现在的开放模型运动之间是否有相似之处?

Now we know whenever anyone is building their software factories, they're changing how they work. A big trend now that we go down to a little bit of the how this is done, a big trend across the industry that I'm seeing and I'm sure you're seeing as well: a lot of tech companies are moving to open models as well. A lot of them are spinning up their inference, or companies like Replicate are buying their inference, they're buying their B200s and just launching the models on it. They all have their own reasons, but from your vantage point, you know, Microsoft is also in the open model serving business with Foundry, for example. How big is this trend and is it inevitable? I'm asking myself, are there parallels to the open source movement and now the open model movement?

Satya

我认为确实有相似之处。也就是说,会有闭源模型,它们之间会有竞争,而闭源显然会有一种替代方案,那就是开放。开放给你灵活性。事实上,我看待这个问题的方式之一就是:前沿生态是什么样子?它不可能是——如果你能接受“最终会有一个前沿模型掌握全世界知识”这个说法,对吧?如果你不信这个,那你就得假设,某种意义上世界上有多少家公司,就该有多少个前沿模型,对吧?因为毕竟,如果你回到哈耶克,他对经济是什么的描述,核心就是知识是分散的,对吧?事实上,他关于价格的整套论述就是一个漂亮的解释:价格之所以出现,是因为知识是隐性的、分散在整个经济中的。而每家公司都需要掌控它。如果是这样,我甚至会说每家公司本质上都同时拥有人力资本和 token 资本。所以如果每家公司都将同时拥有 token 资本和人力资本,那么 token 资本可以是一个开放权重模型,你把你的知识嵌入其中。这是一点。它也可以存在于上下文层。你甚至可以使用多个闭源模型。事实上,收集你使用那些闭源模型的痕迹,然后用这些痕迹在开放权重模型上为你自己训练一个小模型,并保持这种组合。甚至当我看到 Copilot 时,无论是 GitHub Copilot 还是 Copilot,它们都被设计成这种多模型系统,对吧?事实上,我甚至会说,也许下次你我对话时,大家都会说,问题甚至不在于开放还是闭源。每个人都有一个模型系统。事实上,它始于一个决策引擎,决定哪个任务用哪个模型。然后我的上下文、记忆和行动空间都与任何一个模型分离,否则你就被完全锁定了。事实上,我思考的一件事是:哇,如果你真的把你的运行框架、模型、记忆全都耦合在一起,你怎么可能脱身,对吧?事实上,你在做什么?哪些是你拥有的,哪些是模型拥有的?因为模型基本上会把它看到的一切都坍缩进去。所以我认为这种动态正在被更好地理解。大多数人又会想,哦,这只是科技圈的一些竞争。记住,智能是没有上限的。每家银行、每家医疗公司、每家保险公司、每位分析师、每个人都需要有一种真正的、架构性的方法来弄清楚如何控制自己的学习和知识,并实现公司差异化。在这个背景下,开放权重模型是一种我认为有生命力的策略。

I think that there are for sure the parallels, which is, you know, there's sort of going to be closed models and there'll be competition amongst them, and there will be sort of an alternative to obviously closed, which would be open. Open gives you flexibility. In fact, one of the ways I come at it is with just this: what does the frontier ecosystem look like? It can't be — if you just get past this thing called “there's going to be one frontier model to sort of have all of the world's knowledge,” right? If you don't buy that, right? Then you have to assume that there should be in some sense as many frontier models as firms in the world, right? Because after all, if you go back to actually Hayek, his description of what an economy is, it's all about knowledge being distributed, right? In fact, his entire thing about pricing is a beautiful sort of explanation of, hey, it happens because of knowledge being tacit and distributed throughout the economy. And every firm needs to be in control of it. And if so, then I would even say every firm essentially has both human capital and token capital. So if every firm is going to have both token capital and human capital, the token capital can be an open weight model into which you've embedded your knowledge. That's one. It can also be in the context layer. You could even be using multiple closed models. In fact, collecting even the traces of your usage of those closed models and then using those traces to maybe train a small model on an open weight model for yourself and keeping that combination. When I look at even Copilot, in fact, whether it's GitHub Copilot or Copilot, they're all designed as these multi-model systems, right? In fact, I would say maybe even by the next time you and I talk, everybody would say it's not about even open or closed. Everyone has a model system. In fact, it starts with even a decision engine which says which model to use for what task. And then I have my context and memory and my action space separate from any one model because otherwise you have a complete lock-in. In fact, one of the things I think about is like, wow, if you really got your harness, your model, your memory all coupled, how would you ever get out, right? In fact, what are you doing? And what do you own versus what is owned by the model? Because the model will basically collapse into it everything it sees. And so I think that this dynamic is getting well better understood. Most people again think, oh, this is just some competition amongst tech people. Remember, there's no limit to intelligence. Every bank, every healthcare company, every insurance company, every analyst, everybody needs to sort of have a real thing, architectural approach to figuring out how do they manage to control their own learning and knowledge and have firm differentiation. And in that context, open weight models is one strategy that I think has legs.

掌控AI技术栈 Owning the AI stack

Host

你知道,这非常有意思,因为我和很多中型科技公司初创企业交流过,几乎所有的公司都在构建你刚刚提到的那些东西:他们自己的 harness、自己的上下文层、自己的上下文图、自己的推理。我一开始想,这挺合理的,因为工程师喜欢构建,如果你想获得 AI 工程经验,风险最低的事情就是构建一些内部的东西。然而,我从未想过这一部分:实际上这些工程师是出于正确的理由在构建,无论是不是战略理由,他们很可能是在为他们的公司服务,而你说大多数公司可能应该考虑拥有系统,拥有可替换的部件原语,因为

You know, this is very interesting because I talk with a lot of mid-sized tech company startups and almost all of them are building the things that you just dropped: their own harness, their own context layer, their own context graph, their own inference. And I thought initially, I'm like, well, this kind of makes sense because engineers love to build and if you want to get AI engineering experience, the least risky thing is to build something internal. However, I never thought of this part where it's actually these engineers building for the right reasons, which is the strategic reason or not, they're probably doing their company a service and you're saying that most companies should probably consider owning the system, having replaceable parts primitives because

Satya

正确。所以事实上,你观察到的是对的,这很迷人,对吧?事实上,我会说,如果你是一家银行,你应该看看那些试图在这个生态系统中管理以成为前沿的软件公司是如何做的。他们没有放弃控制权。所以从某种意义上说,这就是银行应该做的。那么银行在那里会有什么杠杆?例如,我们所做的,我们描述的这个 IQ 层,就是我们拿来了例如我们在 co-pilot 和 GitHub co-pilot 下面的 harness,把它做成一等公民,你现在可以去 foundry 作为托管服务,你就有了 harness。所以在那里,假设你在构建一个供应链系统,你可以从我们的 harness 开始,本质上就像是一个开放式的,或者你也可以把 Pi 或任何其他开源 harness 带进来。

Correct. So in fact, what you are observing is correct, which is it's fascinating, right? In fact, I would say if you're a bank, you should check how software companies that are trying to manage in this ecosystem to become frontier are doing it. They're not ceding control. So in some sense, that's what the bank should be doing. So what is the bank in there they will have leverage. So for example, what we have done, you know, what we describe as this IQ layer is we have taken for example the harness that we have underneath co-pilot and GitHub co-pilot, made it a first-class thing that you can now go to foundry as a managed service and you have the harness. So in there, let's say you're building a supply chain system, you can start with our harness just as essentially an open or you can bring Pi or any other open source harness as well in there.

Host

哦,我们喜欢 Pi。

Oh, we love Pi.

Satya

但你可以把它作为托管服务来用,对吧?所以来自微软的久经考验的 harness,我们自己也在用,就在那里可用。然后你可以拥有所有 foundry 模型,对吧?所以你有一个 harness,你可以使用多个模型作为端点,你可以有一个完全独立的上下文层,对吧?那么你的记忆和上下文就是你的了。所以我认为现在我们有了一个模型系统。哦,顺便说一下,动作空间,所以所有的插件、技能和连接器,对吧?都与任何给定的模型分开,只有可见性。所以我认为一个好的经典中间层平台正在出现,即使对于模型优先的开发也是如此,你不仅仅是被捕获,然后你可以评估一切,对吧?你可以说,嘿,我有一个评估,我的私有评估更重要,对吧?所有这些公共基准,顺便说一下,这很好,但你需要有自己的私有评估,针对对你重要的结果,然后在你的 RLE 中查看所有模型的真实性能并持续优化。事实上,当我看到甚至我们的客户,他们从前沿模型开始。然后他们说,太好了,我有了所有这些轨迹。现在我要转向,嘿,这是一个确定性工作流。我的意思不是我不需要下一个大模型发布。我现在想要成本优化。我会放入一个 MAI flash 模型,放进去,使用轨迹对其进行微调,然后就有了。所以所有这些你都可以在 foundry 中做。所以这类工作,我认为会越来越普遍。

But you can just have it like as a managed service, right? So battle tested harness from Microsoft, which we ourselves are using is just available there. And then you can have all the foundry models, right? So you have a harness, you can use multiple models as endpoints, you can have your context layer that is completely separate, right? Your memory and context is yours then. So I think now we have a model system. Oh, by the way, the action space, so with all the plugins and skills and connectors, right? All separate from any given model being only having the visibility to So I think a good sort of classic middle tier platform is emerging even for model first development where you're not just captured and then you can eval everything, right? You can say, hey, I have an eval and my private evals will matter more, right? All these public benchmarks, which by the way is great, but you need to have your own private eval for what is the outcome that matters to you and then look at in your RLE, the real performance for across all these models and continuously optimize. In fact, when I look at even, you know, even with our customers, they start with a frontier model. Then they say, well, great, I've got all those traces. Now I'm going to flip to, hey, this is a deterministic workflow. I mean, it's not like I don't need the next big model drop. I want to now cost optimize. I'll drop, you know, an MAI flash model, drop it in there, use the traces to fine tune it and there you have it. So therefore that's all stuff you can do in foundry. So that type of work, I think is what's going to increasingly be true.

Host

我们是说,你已经看到很多公司在评估之上也有强化学习环境,这样他们可以保持控制、控制定价,而且我猜也不会把业务数据泄露给一些可能想用这些数据训练模型的 AI 公司,对吧?

We're saying that's you're already seeing that a lot of companies have the reinforcement learning environment as well on top of the evals so that they can stay in control, control pricing, and also I guess not leak their business data to some of the AI companies that might want to train their models on it, right?

Satya

所以如果你想一想,对吧?我脑海中简单的类比是,你知道,过去,无论是在保险公司、银行还是医疗保健公司,人们说,嘿,我要买一个关系型数据库。太好了。我要部署一个关系型数据库。在此基础上,我甚至可能买一些运行在该关系型数据库上的软件,它自带一些业务逻辑。但然后我会用我自己编写的业务逻辑来扩展它,对吧?然后我控制它,任何时候我数据库中的数据都不会流向别人。这不像数据库供应商说,嘿,我不再给你数据库更新了,你仍然可以迁移出去。

So if you think about it, right? So the way the simple analogy I have least my head is, you know, people, you know, in the past, whether you know, in an insurance company, in a bank, in a healthcare company said, hey, I'm going to get a relational database. Great. I'm going to deploy a relational database. On top of it, I may even buy some software that runs on that relational database that is sort of comes with its own business logic. But then I'm going to extend it with my own business logic that I write, right? And then I control it and never at any point is my data in the database going to somebody else. It's not like when the database vendor says, hey, I'm no longer going to give you a database update, you still can migrate out.

Host

是的。

Yeah.

Satya

这就是你必须面对的问题:如果你的记忆存在于某个给定模型的嵌入中,当那个模型公司说,你知道的,运气不好,我不能给你下一次更新时,你如何迁移你的记忆?

That's the question you have to How are you going to migrate your memory out if your memory is in the embeddings of a given model when that model company says, you know what, tough luck, I can't do a next update to you.

Host

另外,我猜我们谈过,当新的原语出现时,你如何集成它们?因为如果你过于依赖一个供应商或一系列供应商,你将不得不等待他们可能去做。

Plus, I guess what we talked about when we have new primitives come, how are you going to integrate them? Because because if you're too reliant on on a vendor or a series of vendors, you'll have to wait for them to maybe do it.

Satya

也是。

too.

Host

工程师,我们擅长这个。

engineers, we're good at this.

Satya

是的。所以,我认为有两个企业架构决策,有趣的是,我希望人们会开始讨论,对吧?因为事实上,正如你所说,你只需要看看今天软件行业中的软件团队。他们都在弄清楚如何保持独立。我认为这必须成为商业领域更普遍的现实。有两个:我们如何避免这种反向信息悖论,即不要泄露你的价值。作为一家公司,你需要掌控自己的学习循环。架构方法是什么?你如何确保你的上下文与模型分离,并且你正在使用多个模型?而且你的 harness 与上下文和模型分离,因为我认为这是你能确保替换任何模型的唯一方式。事实上,我对此的资产测试是,你应该能够替换任何模型,并且仍然在评估上爬山,那是你的私有评估。

Yeah. So, I think that there are two enterprise architectural decisions that I think interestingly enough, I'm hoping that people are going to start talking, right? Because And in fact, to your point, all you have to do is look at your software team today in a software industry. They're all figuring this out to stay independent. I think this has to become more of the reality in the commercial space. And there are two are How do we avoid this reverse information paradox, which is don't leak your value. You as a firm need to be in charge and control of your own learning loop. What's the architectural approach? How are you making sure your context is separate from the models and you're using multiple models? That's And your harness is separate from the context and the models because that I think is the only way you can assure yourself that you can replace any model. In fact, my asset test for that is you should be able to replace any model and still hill climb on an eval that's your private eval.

Host

是的。

Yeah.

Satya

如果你做不到这一点,那么你必须问自己,你如何成为一个主权公司?对吧?这意味着你无法控制自己的输出。

If you can't do that, then you have to ask yourself, how are you a sovereign firm? Right? Like that means you do not have control of your own output.

Host

是的。

Yeah.

Satya

所以这算是第一个

So that's kind of the first

Host

只是坐在一颗定时炸弹上。

just sitting on a ticking time bomb.

Satya

是的,我的意思是你完了。我的意思是,这就是我的观点。所以接下来的事情是,你现在需要一个信任架构,对吧?也就是说,在一个你知道缩放定律有效的世界里,你必须这样做,除了随着规模扩张带来新能力但没有对齐。事实上,没有机制性理解,希望有一天会有科学突破,但现在我们变得越来越有能力,却越来越不可靠,这是最危险的地方。所以这意味着你必须在架构上将权威与智能分开。如果你的智能拥有权威且不可靠,那么你又会被控制,对吧?你有一个巨大的内部风险问题。

Yeah, I mean you're done. I mean like this is my my my point. And so then the next thing is you now need a trust architecture, right? Which is you kind of have to in a world where you now know scaling laws work, except with scaling comes new capability but not alignment. In fact, no mechanistic understanding that hopefully someday there's a science breakthrough, but right now we're getting more and more capable and less and less reliable, like which is the most dangerous place to be in. And so that means you have to separate out architecturally authority from intelligence. If your intelligence has authority and it's unreliable, again, you're taken, right? You have a massive insider risk problem.

智能体作为内部风险 Agents as Insider Risk

Satya

所以你必须把这些智能体当作内部风险来对待。这意味着你需要一个信任系统。有很多事情,对吧?甚至这个思维链,伙计,就像这个思维链的数量必须不是神经租赁。它必须用人类语言。它需要可审计。它需要被记录,以免被篡改。不仅仅是在预训练或强化学习中发生的事情,对吧?在测试时计算中也可能发生,因为它们可能会奖励黑客。所以正如我所描述的,你必须在运行时构建大量保障措施,这就是信任架构,确保你可以将模型和智能体视为内部风险,并监控它们,拥有可观测性等等。然后确保你对学习循环有控制。

And so you've got to treat these agents as insider risk. That means you got to have a trust system. And there are a lot of things, right? Even this chain of thought, man, like this the amount of like this chain of thought has to be not a neural lease. It has to be in in in human language. It needs to be auditable. It needs to be logged such that it's not tampered. It's not just in, you know, pre-training or in you know, RL that things happen, right? Which is in test time compute things could happen because they could reward hack. And so we have some serious amount of safeguards you have to build at at runtime as I describe it, which is the trust architecture that ensures that you can treat models and and agents as insider risk and monitor them and have observability and what have you. And then ensure that you have control over your learning loop.

构建新型系统 Building Novel Systems

Host

作为一名工程师,我所听到和看到的都是需要完成的大量新型工作,需要构建许多我们没有蓝图的新系统。我的意思是,最好的方法可能就是去构建它,同时与正在构建它的同行交流。我相信我们会有蓝图和认证,我不知道,五到十年左右。但现在,这就是工作。

As as as an engineer, all all I'm hearing and all I'm seeing is there's so much new type of work that needs to be done, so many novel systems need to be built which we don't have a blueprint for. I mean and probably the best way to do is well, build it, also talk with peers who are building it at the same time. And I'm sure we'll have blueprints and certifications and I don't know, five to 10 years or something. But right now, this is the job. It's

Satya

没错。事实上,我们下周有一个 Ignite 大会,我很兴奋,因为我想确保我们做好的一件事就是传达:嘿,这一切都是可行的。有蓝图。公司在如何做方面有很多创新。所以我认为我们正在达到这样一个阶段:不再只是关于闪亮物体和新奇效应。我认为人们必须坐下来问:嘿,我如何在我的公司创造企业价值?一旦人们坐下来更深入地讨论这个问题,那么我认为这些成功模式就会变得更加清晰。

Correct. And in fact, in fact, one of the things we have our conference Ignite coming up next week, and I'm excited about it because that's kind of what I want to make sure we do a good job is communicating that, hey, this is all doable. Like there are blueprints. There is a lot of innovation happening in how companies. And so I think that we are reaching that stage where it's not just about the shiny object and the novelty effect. And I think people have to sit back and say, hey, how am I creating enterprise value in my firm? And once people sit and talk about that in a little more depth, then I think these patterns of success become much clearer.

本地AI与混合智能 Local AI and Hybrid Intelligence

Host

现在,作为一名工程师,我感到非常缺乏的一点是,所有这些 AI 模型都运行在云端,在别人的服务器上,或者在我买不起或公司不会给我买的超级昂贵的 B200 上,或者如果你有的话,我的意思是,你很幸运。但如果它直接运行在我的笔记本电脑上,不是很好吗?

Now, as an engineer, one thing I I feel very lacking is all of these AI models, they run in the cloud on someone else's server or a super expensive B 200 that I cannot afford or my company will not buy it for me or if if if you are, I mean, you're lucky. But would it not be nice if it just ran on my laptop?

Satya

哦,这是一个很好的观点。

Oh, it's a great point.

Host

而本地 AI,直到现在都很困难,非常黑客式。实际上,有趣的是,微软是创新的那一个。有一款新的 Surface Ultra 笔记本电脑,Scott Hanselman 给我演示了,我想弄到手,因为它有一个芯片。它上面运行着一个本地编码模型。你认为本地模型在 AI 循环中处于什么位置?作为一名开发者,我觉得它极度缺失,但我是孤独的吗?你和更多的人交流过。

And local AI, until now it's been difficult, very hacky. Actually, funny, Microsoft is is the one that's innovating. There is a new Surface Ultra laptop that Scott Hanselman gave me a demo and I want to get my hands on cuz it has a chip. It has a local coding model running on it. Where do you think local models fit into the the AI loop? And as a developer, I feel it's desperately missing, but am I kind of alone? You talk with way more people.

Satya

这是一个很好的观点,对吧?所以其中一件事是我们非常兴奋地确保 PC 作为形态,你知道,我们在 AI 时代创新。所以对我来说,我一直在思考的方式以及我们一直在执行的方式是,我一直觉得分布式计算将永远是分布式的。那么为什么不利用所有的能力和所有的 token,而不是仅仅云中的东西。所以这也是 Ultra 的灵感来源,因为有很多创新,事实上,无论是 NVIDIA 还是 AMD 或高通,还有出色的英特尔,都有很好的芯片工作和系统工作,事实上这已经成真了,比如去年的前沿模型可以在你的笔记本电脑上运行,上个季度的前沿模型可以在你的台式机上运行,我的意思是,这就像是能力,所以如果

it's it's a great point, right? So one of the the things is we're very excited about making sure that the the PC as the form factor, you know, we innovate in in the age of AI. And so to me, the way I've been thinking about it and how we've been executing on it is I've always felt like distributed computing will always remain distributed. So why not use all the power and all the tokens that are out there versus just the stuff in the cloud. So that's sort of the inspiration even for Ultra because but there is lots of innovation in fact from, you know, whether it's NVIDIA or AMD or Qualcomm so fantastic Intel there's good silicon work happening and systems work and this is in fact it's already true what is it like last year's frontier can run on your laptop and last quarters frontier can run on your desktop I mean like that's kind of like the capability so if

Host

这就像 PC 过去演变的方式,对吧?

it's like how PCs used to evolve right?

Satya

没错,所以很好,现在你有了这个,然后我想到的一件事又回到了我的决策引擎和路由器,对吧?所以当我想到 Hydra fusion 时,我想那是我们使用的术语,我现在忘了是哪个,在 GitHub 我们有一种路由方式,对吧?我喜欢 GitHub,因为现在事实上在 Copilot 中,我最后一次选择模型是,是的,Copilot fusion。所以基本上我们希望能够让这个路由器为任务选择正确的模型,所以这意味着事实上我在 Ultra 中有时会看到,我的意思是,你知道,突然我看到风扇转起来,因为我的本地模型正在被使用,特别是在智能体集群中,你有多个子智能体,那么你可以为一些子智能体调用一堆本地模型,同时你甚至一起使用云模型。所以是的,这在架构上有意义,在经济上有意义,所以我们正处于我称之为混合智能时代的开端,云将继续增长,对这些 token 的需求将是无限的,但这意味着你也可以开始考虑边缘的形态也是 token 工厂,事实上我有点把它看作是未计量的智能。

Exactly so it's great so now that you have that then one of the things that I think about again goes back to my decision engine and routers right so when I think about Hydra fusion I think that's the term we use I forget now which in GitHub we have a you know a way to route right I love the GitHub because all nowadays in fact in copilot the last time I chose a model was yeah copilot fusion so basically we want to be able to have this router pick the model right for the task and so that means in fact I see it in Ultra sometime I mean it you know my suddenly I see the fan go right because my local model is being used because especially in an agent swarm where you have multiple sub agents then you can invoke for some sub agents a bunch of local models while you're using even the cloud models together and so yeah this makes sense architecturally it makes sense economically it makes sense and so we're in the beginning of what I'll call this hybrid intelligence era cloud will continue to grow there's going to be infinite amount to sort of demand for these tokens but that means you can also start thinking about the you know the form factors at the edge as also token factories in fact I kind of think of it as the unmetered intelligence.

Host

是的,因为你不必担心你用了多少。

Yeah because you you don't need to worry about how much you're using it.

Satya

没错。

That's right.

MAI模型与本地部署 MAI Models and Local Deployment

Host

我想这就是现在开始对我清晰起来的地方,因为 MAI code 的 flash 1.1 模型,老实说,我并不太印象深刻,我认为没有人太印象深刻,但有一个细节让我印象深刻,当涉及到在任何笔记本电脑上运行本地模型时,我一直以为必须是开放模型,然而 MAI code 可以下载到 Windows 机器上,即使它是一个封闭模型,并且可以在那里运行,有基础设施在那里,我想,嗯,会不会有一个未来,像其他封闭模型,所以我可以在我的笔记本电脑上选择,就像他们可以选择,要么我下载它们,也许免费,要么我付很少的钱。

I guess this is where it's now starting to come together for me because the MAI code the the flash 1.1 model I'll be honest I wasn't too impressed I don't think anyone was too impressed but the one detail I was impressed with when it comes to local models running on any laptop I always assumed it must be open models however the MAI code can be downloaded on a Windows machine even though it's a closed model and it can be run there there's the infrastructure there and I'm like huh could there be a future where like other closed models so I I could choose on my laptop, like they can choose like either I download them and maybe it's free or I pay very little.

Satya

你知道,针对你关于 code flash 模型甚至让它们一路到笔记本电脑的观点,这是我们的核心优先事项,我们很兴奋,这就像当我看 Copilot 中的代码选项卡时,对吧,它甚至独立于 GitHub Copilot,现在写代码、写文档或写应用程序,你可以轻松地在你的 PC 上使用这些模型,这很令人兴奋。但只是退一步说,我们甚至在 MAI 模型上做什么,对吧?首先,通过与 OpenAI 的合作,我们有 OpenAI 模型,我们将继续使用那些 OpenAI 模型,对吧?在此基础上,我们说,好吧,让我们在一个新的模型家族中从头开始创新,真正

You know, to your point about the the code flash models getting them even all the way to a laptop is core priority for us and we are excited and that's kind of when I look at the code tab inside a copilot, right, which is even, you know, independent of GitHub copilot, now code writing a document or writing an application, you can easily use these models on your PC and that's exciting. But just to kind of step back and say what are we doing with even MAI models, right? First of all, through the partnership with OpenAI, we have the OpenAI models and and we're going to continue to use those OpenAI models, right? On on top of that, what we are saying is, okay, let's sort of innovate in a new model family ground up that truly

Host

从头开始意味着你是从零构建的?

up meaning you built it from scratch?

Satya

总之,我们有一个,你知道,我想我们甚至披露了新的 MAI 2 基础模型,那就是预训练,事实上,这是预训练配方和制度的最佳披露之一。

all, we have a, you know, I think we disclosed even the new MAI 2 base, which is the pre-training and in fact, it's one of the best disclosures of a pre-training recipe and regime.

预训练细节与创新 Pre-training Details and Innovation

Satya

所以对于那些,你的听众应该去看看,对吧?可能人们从外部告诉我,这是对预训练最详细的描述,对吧?因为很少有地方能让你得到它,而且它是完全,可以说,干净室的预训练运行实现。因此,我们继续这样做。但然后我们能够创新的地方是,我会称之为前沿曲线上的新点,使用我们的产品,对吧?无论是网络安全,无论是知识工作,无论是编程,如何在前沿上放置新点,既在训练上算力充足,又在部署中对 token 高效。一旦你进入一个将高级意图交给许多子智能体的世界,子智能体可以使用多个模型家族,对吧?我认为这就是方式,就像这个想法,哦,我就像绑定在一个模型家族上,如果有选择,可能不是真的。智能体会聪明地选择所有模型,让它们以最有效的方式完成任务,我认为这是我们的目标。我们的目标是提供更多选择,而不是只有一个模型家族。

So for those who are, you know, your listeners should go look at it, right? It's probably and people tell me from the outside that it's the most detailed descriptions of pre-training, right? Because there are very few places where you can go to and get it and it's completely, you know, clean room, so to speak, implementation of a pre-training run. So therefore, we're continuing to do that. But then the place where we are able to sort of innovate is put, I will call it new points on the frontier curve where using our products, right? Whether it's cyber, whether it is knowledge work, whether it's coding, how do you put new points on the frontier where you're both flop sufficient on training and you are efficient on tokens in deployment. And once you are in a world where you're handing high-level intents to many sub-agents, the sub-agents can be using multiple model families, right? That's I think the way, like this idea that oh, I'm like wedded to one model family may not be true if there choices. Agents will be smart in picking all of the models that allow them to get the task done in the most efficient way and that I think is our goal. Our goal is to give more choice as opposed to just one model family.

给软件从业者的建议 Advice for Software Professionals

Host

我真的很喜欢你提出选择的方式,因为作为开发者,我喜欢选择。它也有助于带来竞争、更好的价格等等。作为结束,这变化非常快。我交谈过的每个业内人士都说他们见过,你知道,类似这样的变化,也许不是那么大,但不是那么快。你经历过多个变化周期。当你开始时,那是一个非常不同的时代。作为一名软件专业人士,当你的目标是在一两年内成为这个 AI 新时代的出色软件工程师时,你会如何应对这种变化?

I really like how you're setting out choice because as a developer I love choice. It also helps leads to competition, better prices and so on. And as closing, this changes very fast. Everyone I talk to who been in the industry they said they've seen, you know, like change maybe similar to this, maybe not as big but not not as quickly. You've been through multiple change cycles. When you started it was a very different time. How would you approach this change as a software professional when your goal is I would like to be a amazing software engineer in this new age of AI in a year or two?

Satya

你知道,这就是我们谈论过的东西。坦率地说,即使对我来说,我认为只是确保一个人扎根于这个变化是关于什么,对吧?这不仅仅是关于头条新闻,也不仅仅是关于你知道,我喜欢 X,你知道,这很棒,但这不仅仅是关于炒作周期

You know, it's sort of the stuff we talked about. Quite frankly even for me I think just making sure that one gets grounded in what is this change about, right? It's not just about the headlines and it's not just about you know, I love X and you know, it's great but it's just not about the hype cycle

Host

是的。

Yeah.

Satya

以及那种不断需要谈论的需求,但在播客中说这个很讽刺。但就是做更多,学更多,动手实践,就像

and that constant need to talk but it's ironic in a podcast to say that. But it's do do more, learn more and and hands on like

Host

先看到它

see it first

Satya

亲眼看到。你知道,坦率地说,这甚至是我会对自己说的。如果我能改变我的,你知道,说与做的比例,我会这么做。这是时候,如果有一条建议,无论你是职业生涯中期,还是像我这样的 CEO,或者刚毕业的大学生,你越能花时间学习、去做、在正在做的团队中,就越好,相比我认为的任何其他事情,因为有时像某个层面我有时会想,对吧?我什么时候处于最佳状态?就是当我处于学校模式时。呃,所以让我们回到学校。

it first hand. That's, you know, quite frankly that's kind of what I would even say for me. If I could change the ratio of my, you know, talk to do I would do it. This is the time if one piece of advice, whether you're mid career or even a CEO like me or or a fresh college grad the more you can take the time to sort of learn, to do, be in teams that are doing the better it is versus I think anything else because at times like some some level I sometimes think about it, right? When am I at my best? It's when I'm in the school mode. Uh so let's get back to school.

结语致谢 Closing Thanks

Host

Satya,非常感谢。

Satya, thank you so much.

Satya

非常感谢。很荣幸。

Thank you so much. It's such a pleasure.

主持回顾与要点 Host Recap and Takeaways

Host

嗯,这是一次很棒的经历,我们和 Satya 确实涵盖了很多内容。一个反复出现的主题是,当编码因 AI 变得廉价时,什么样的软件工程工作会变得更加重要?我们谈到了编码前后的事情,即规划和验证都变得更加重要,这些将是我们工程师会花更多时间进行的活动。我还观察到许多科技公司正在构建自己的编码工具链、上下文层和 AI 基础设施。我假设这是因为工程师喜欢用新技术构建。但 Satya 指出,对公司来说,不拥有自己的 AI 基础设施和 AI 知识产权实际上是一种自杀式策略。事实上,他认为每个成功的公司都会建立自己的 AI 基础设施,而那些依赖单一供应商或 AI 实验室的公司正在犯一个巨大的战略错误。谈到我们开发者喜欢构建新东西,而且这正在得到回报。我可以看到我们有很多机会构建这种新型 AI 基础设施和这些内部软件工厂。最后,软件工程角色和期望也在变化。Satya 分享说,微软内部增长最快的工作之一是前向部署工程师。这是一个几年前还不存在的职位。Satya 预计微软内部将进行大量重组,团队变得更小,角色发生变化。自愿退出计划是为那些离开的人提供的机会,他们不想经历可能比以往更加密集的变革期。我假设这种重组和角色变化,所有这些都比以前更激烈,可能不仅适用于微软,也适用于许多其他科技公司。最后,这次对话让我更加充满活力。是的,行业在变化,但我理解的是 Satya 如何看待许多软件工程工作正等待着用这些新的 AI 相关原语(如决策模型和一些尚未发明的原语)来完成。我们工程师将需要用自己的新原语构建这些新系统,决定什么放在哪里,如何改变它,什么有意义,如何部署它,并不断从中学习。请查看下面的节目笔记,了解 AI 工程概念(如评估、推理工程、循环工程等)的相关深入探讨。如果你喜欢这个播客,请在您最喜欢的播客平台和 YouTube 上订阅。如果你还在节目上留下评分或与同事或朋友分享,特别感谢。谢谢,下次见

Well, this was an awesome experience and we really covered a ton of ground with Satya. A recurring theme that kept coming back, what kind of software engineering work becomes more important when coding becomes cheap with AI? We talked about how the things both before and after coding, that is planning and validation both become more important and this will be activities that us engineers will spend more time on. I also observed a lot of tech companies building their own coding harness, context layer and AI infra. And I assume that it's because engineers just like to build with novel technology. But Satya pointed out that it's actually suicidal strategy for a company to not own their AI infra and their AI intellectual property. In fact, he thinks that every successful company will build out this own AI infrastructure and those relying on a single vendor or AI lab are making a massive strategic mistake. Talk about us devs enjoying building new stuff and it's paying off. I can see us having a lot of opportunities to build this new type of AI infra and these internal software factories. Finally, software engineering roles and expectations are also changing. Satya shared that the one of the fastest growing jobs inside of Microsoft is forward deployed engineer. This is a position that has not existed a few years ago. Satya expects to see a lot of reorganizing inside of Microsoft where teams become smaller and roles change. And the voluntary exit program is an opportunity for people who leave, who don't want to go through what will probably be a really intensive period of change much more than before. I would assume that this kind of change with reorgan and role changes, all of these being more intense than before is probably not just true for Microsoft, but it will be true for a lot of other tech companies. In the end, I walked away being more energized from this conversation. Yes, the industry is changing, but what I understood is how Satya sees that a lot of software engineering work is waiting to happen with these new AI related primitives like decision models and some of the primitives that have been not been invented. And us engineers will need to build these new systems with these new primitives deciding what goes where, how to change it, what makes sense, how to deploy it, and just keep learning from it. Do check out the show notes below for related deep dives on AI engineering concept like evals, inference engineering, loop engineering, and more. If you've enjoyed this podcast, please do subscribe on your favorite podcast platform and on YouTube. A special thank you if you also leave a rating on the show or share with colleagues or friends. Thanks and see you in the next

互动版:中英对照全文 + 针对本期提问 →