The New Physics of Business — Garry Tan, Y Combinator

频道: AI Engineer
视频: https://www.youtube.com/watch?v=eBUyTS7SzV4
原文语言: en
统计: 共 17 轮 · Garry 17

[0:01] Garry

[music]

（音乐）

[0:13] Garry

Okay, great. Hey everyone. How's everyone doing? All right. Are we ready for the revolution? Okay, Theo just asked the right question. What do we build now? I'm going to answer it from the other side of the table. Uh I'm a founder, I'm an investor, and I run a 20-year-old institution that is becoming AI native right now, which is a strange and wonderful thing to do to a 20-year-old institution. Uh and I'll spend about 20 minutes um talking about, you know, what YC is we're trying to build companies where one person does what it took to two it One person does what used to take a thousand people. And I don't mean that as a metaphor. I mean that me- mechanically this year, the people in this room will do this. In about an hour, some of you will walk into the startup battlefield, and I want you to walk in knowing what's actually possible right now, because what is possible now is much much bigger than what people believe. So, let me start with a number and uh you know, I got torn apart on the internet for this, but I'm going to say it again in front of all of you anyway. Uh this is the one room in the world that will stress test it, and I'd rather stress test it with you myself. In 2013, I was a YC partner uh building the internal social network um at at YC. Uh I was also investing in companies, but you know, I was also uh near full-time engineer. Um and when I was doing that, I could maybe do about 14 usable logical lines of code a day. Take out all comments, take out all the and that's how many lines of code I I was writing. And if you look at the literature from that era, um you know, that's kind of normal. Like some people write 15, some people write 50. It was not, you know, the thousands of lines of code that I know a lot of you in this room are actually writing now per day. Um that's about median 15.

好的，太棒了。大家好，各位今天怎么样？好，准备好迎接这场革命了吗？刚才 Theo 问了一个非常关键的问题：我们现在该做什么？我打算从桌子另一头的视角来回答它。我是一名创始人，也是一名投资人，同时我在掌管一家有 20 年历史的机构，而它此刻正在变成 AI 原生（AI native）——对一家 20 岁的机构来说，这是一件既奇怪又美妙的事。接下来大约 20 分钟，我想聊聊 YC 在做什么：我们想打造这样一种公司——一个人干出过去要一千个人才能完成的活。我说的不是比喻，我是说在机制层面上，今年，在座的各位就会做到这件事。大约一小时后，你们中的一些人将走进创业竞技场（startup battlefield），我希望你们进场时清楚地知道当下到底能做到什么，因为现在能做到的，远比人们以为的要大得多。那我先抛一个数字——我因为这个数字在网上被喷得体无完肤，但我还是要当着你们所有人的面再说一遍。这是全世界唯一一个能对它做压力测试的房间，我宁愿亲自和你们一起来测。2013 年，我是 YC 的合伙人，在 YC 内部搭建内部社交网络。我同时也在投项目，但基本算是一名近乎全职的工程师。那时候我一天大概能写出 14 行真正可用的逻辑代码——去掉所有注释、去掉所有杂项之后，就是这个数。如果你去翻那个年代的文献，你会发现这其实挺正常的：有人写 15 行，有人写 50 行。反正绝不是现在这个房间里很多人每天实际在写的成千上万行代码。中位数大概就是 15 行。

[2:18] Garry

That was me at full effort at that time. This year I run uh YC full-time. Uh same person, same hours, actually way less hours weirdly. Uh you know, but I have a 5:00 p.m. kid pick-up now. And I did the math on my output, and it's about 400x. Now, before the skeptic in the third row right there deflates the number from for me, let me deflate it myself. If you don't trust the raw code, well, fine. Take the most pathological verbosity penalty you can stomach, and assume the agent writes bloated code. Assume half of it is scaffolding. Assume I'm flattering myself. It's still 8x at the floor and 80x in the middle. That number is large, no matter how you torture it. And here's the part that matters, the part that I'd tattoo on the inside of everyone's eyelids if I could. It's not the model. The 2x people and the 100x people are using the exact same Claude, same weights, same context window, same API. So, the leverage is not in the weights. It's in how you wire the work. And it's not just me. At YC, we see this all the time. In the winter 25 batch, a quarter of the companies had code base code bases that were 95% AI-generated, and that was a year ago. That batch has become the fastest-growing, most profitable batch in the history of YC. 94 companies total have now crossed a hundred million in dollars in revenue from a seed check in the history of YC. So, uh I think we know what we're talking about here. And I can't prove that the AI-generated code caused the growth, but what I can tell you is the fastest growing founders we fund are not treating AI as auto complete. They're treating it as a workforce. The companies that wired the work differently are the ones that are bending the curve. So, what does wiring the work really mean? This is the heart of the talk. And this is what I most want you to steal.

那就是当年火力全开的我。今年，我全职运营 YC。同一个人，同样的工时——其实说来奇怪，工时还少了不少，因为我现在下午五点得去接孩子。我算了算自己的产出，大约是 400 倍。在第三排那位怀疑论者帮我把这个数字戳破之前，我先自己来戳。你要是不信原始代码量，行，那就套上你能忍受的最苛刻的『冗余惩罚』：假设 agent 写的代码很臃肿，假设一半都是脚手架，假设我在自吹自擂——就算这样，下限也有 8 倍，中间值 80 倍。不管你怎么折磨这个数字，它依然很大。而真正重要的部分来了——如果可以，我恨不得把这句话纹在每个人的眼皮内侧：关键不在模型。那些 2 倍的人和 100 倍的人，用的是一模一样的 Claude，同样的权重、同样的上下文窗口、同样的 API。所以杠杆（leverage）不在权重里，而在于你如何组织工作（wire the work）。而且不只是我一个人这样。在 YC，我们经常看到这种现象。2025 冬季批次里，有四分之一的公司代码库 95% 是 AI 生成的——那还是一年前的事。那个批次已经成为 YC 历史上增长最快、最赚钱的一批。YC 历史上，一共有 94 家公司从一张种子轮支票起步，年收入跨过了一亿美元大关。所以，我想我们很清楚自己在说什么。我没法证明是 AI 生成的代码导致了这种增长，但我可以告诉你：我们投的那些增长最快的创始人，没有把 AI 当成自动补全（autocomplete）来用，他们把它当成一支 workforce。那些以不同方式组织工作的公司，正是那些把增长曲线掰弯的公司。那么，『组织工作』到底是什么意思？这是这场演讲的核心，也是我最希望你们偷走的东西。

[4:16] Garry

Everything we've learned building with agents maps to an organization. Sorry, there's no slides.

我们在用 agents 做开发的过程中学到的一切，都能映射到一个组织上。抱歉，我没有幻灯片。

[4:24] Garry

[laughter]

（笑声）

[4:26] Garry

I have no slides. I'm so sorry.

我真的没有幻灯片，非常抱歉。

[4:29] Garry

[applause]

（掌声）

[4:31] Garry

A skill file is an employee. It has one capability, one job, written down clearly enough that someone can execute it. Uh a resolver table, uh the thing that many of you, you know, when you run into cloud code and it says your context is too big in cloud.md, you run off and create a resolver table. Well, you know, it and if you don't know what that is, it's literally like whenever you need to uh alter a test, load test.md, you have a whole table of these things. Um that's an org chart. A task comes in and the resolver decides who handles it and where it goes. Uh filing rules are your internal process. So, this can be um you know, whether or not the resolver is actually working and uh is you know, is there a is it actually in compliance? And uh trigger e-vals. So, going in and actually having a test that says, "When I need to alter a test file, does test.md actually get loaded?" Those are performance reviews. So, you know, what have we done? Like literally every part of an organization, the organization that you used to have to hire a thousand people for, I just told you what those things are. They're markdown files and other types of markdown files. And maybe there's some TypeScript in there, too. We've been building organizations this whole time, but we didn't have a management layer, but now that's what we have. When you sit down with Claude Coder Codex, you're not writing software, you're hiring, training, and managing a workforce made of markdown. Uh and you know, that's there are tons of companies that I see that are doing this. Uh Emergence, an AI app builder out of Summer 24, they went from public launch to nine figures of ARR in eight months. Uh when they crossed $15 million ARR, uh they were only 15 people. Retail out of Winter 24, it's at $60 million with about 40 people. The kind of that kind of revenue per head did not exist before. Not in software, not in oil, not in railroads, never.

一个 skill 文件就是一名员工。它有一项能力、一份职责，写得足够清楚，清楚到任何人拿到都能执行。再说 resolver table（解析表）——你们很多人都遇到过：跑 Claude Code 的时候，它提示你 CLAUDE.md 里的上下文太大了，于是你就跑去建了一张 resolver table。如果你不知道那是什么，它其实就是一张对照表，比如『每当需要改测试时，就加载 test.md』，你会有一整张这样的表。那玩意儿就是组织架构图（org chart）：一个任务进来，resolver 决定谁来处理、该派到哪里去。而 filing rules（归档规则）就是你的内部流程——比如判断 resolver 是否真的在正常运转、是否合规。还有 trigger evals（触发评估）：真的去写一条测试，验证『当我需要改测试文件时，test.md 是不是真的被加载了』——这些就是绩效考核（performance reviews）。所以你看，我们做了什么？一个组织的每一个组成部分——那个你过去得雇一千个人才能撑起来的组织——我刚刚已经告诉你它们都对应什么了：它们就是一堆 markdown 文件，外加另一些 markdown 文件，也许里面还夹了点 TypeScript。我们其实一直都在搭建组织，只是过去没有管理层（management layer），而现在我们有了。当你坐下来打开 Claude Code 或 Codex 的时候，你不是在写软件，你是在招聘、培训和管理一支由 markdown 组成的 workforce。我看到大量公司正在这么干。Emergence，Summer 24 批次的一家 AI 应用构建公司，从公开发布到年收入九位数（ARR）只用了八个月；当他们跨过 1500 万美元 ARR 的时候，团队只有 15 个人。Winter 24 批次的 Retail，现在做到了 6000 万美元，团队大约 40 人。这种人均收入水平以前根本不存在——软件行业没有，石油行业没有，铁路时代也没有，从来没有过。

[6:38] Garry

These are not freaks of nature. They're just the first companies built natively on the new physics. And so, how do companies like that actually run? Not by hiring hundreds of people for sales, support, ops, and finance. The AI-native companies I see inside YC encode all of that as skills, written procedures that their agents execute, and they hire they hire engineers whose job it is to maintain those skills, to do the work the skills can't do yet. That is an AI-native company, and it's not a thought experiment. It'll actually file your taxes if you have a skill file for it. Now, picture YC's batch room. Like it actually kind of looks like this. Um 400 companies or 400 founders at long tables, and uh you know, I can imagine every single one of you each with a laptop every single day. You're doing uh a former person's entire year worth of work in a single day. That's not the future. That's actually the bar right now. And if you're not doing it, your competitor is, and they will eat your lunch politely and thank you for it. Here's the extension uh most engineering stocks miss. It's not just the engineers. At YC, as we make our transformation, it's our media people, our event staff, our finance team. People have never opened a terminal in their lives are building skill files and cron jobs. And one of our finance folks just collapsed about a hundred Excel workbooks into a single app she built with our internal open claw and company brain. She's not a programmer. She's a manager of agents now, and everyone at YC now is. So, that's why YC can run at the scale it does with a staff that would look like a rounding error at any other comparable firm. And that's not because we work harder, because it's because we have a different type of org. And that's whole the whole game. It's not just 400X engineers. It's one company that operates at the level of 400X everyone else.

这些公司并不是什么天赋异禀的怪胎，它们只是最早一批原生建立在这套『新物理法则』之上的公司。那么这样的公司到底是怎么运转的？不是靠为销售、客服、运营和财务雇上几百号人。我在 YC 内部看到的那些 AI 原生公司，把这一切全部编码成 skills——写成书面流程，由他们的 agents 来执行；然后他们雇工程师，工程师的工作就是维护这些 skills，并去做 skills 暂时还干不了的活。这就是一家 AI 原生公司，而且这不是思想实验——只要你为它写好 skill 文件，它真的能帮你报税。现在，想象一下 YC 的 batch 教室——它实际看起来还真差不多是这样：400 家公司、400 位创始人坐在长桌前。我完全可以想象你们每一个人，每天抱着一台笔记本，一天之内干完过去一个人一整年的工作量。这不是未来，这就是当下的及格线。如果你没在这么干，你的竞争对手正在这么干，而且他们会礼貌地抢走你的午餐，还顺便谢谢你。接下来是大多数工程类演讲会漏掉的延伸：这不只关乎工程师。在 YC 推进转型的过程中，我们的媒体团队、活动人员、财务团队——那些一辈子没打开过终端的人——都在写 skill 文件、建 cron 任务。我们财务团队的一位同事，刚刚用我们内部的 Open Claw 和 company brain 把大约一百个 Excel 工作簿压缩成了她自己搭的一个应用。她不是程序员，她现在是一名 agents 的管理者——如今 YC 的每个人都是。这就是为什么 YC 能以这样的规模运转，而员工人数放在任何同等规模的机构里都小到像个舍入误差。这不是因为我们更拼命，而是因为我们拥有一种不同类型的组织。这就是整场游戏的关键：不只是 400 倍的工程师，而是一整家公司，以 400 倍于所有人的水平在运转。

[8:36] Garry

And so, you know, if you remember only one thing about this, I mean, this is one of the things I had to discover along the way. Like, you actually have to be really, really careful about where the compute computation is actually happening. It's happening almost always in two different places. And all of the bugs, all of the AI engineering that we run into that's a problem, it's usually because something is happening in one side of the equation that should be in the other. Uh the first area I would say is latent space. So, the actual LLM. It's you know, what do you use it for? Taste judgment, understanding what a human actually wants when they say something vague, the non-deterministic calls, the computation that lives in the model, and you steer it with the markdown file. And then deterministic space is what engineers know. Like, your your code agents go off and write TypeScript or, you know, maybe they're writing Erlang if you're using Elixir. Um Yeah. Deterministic space is the second place. Let's say, you know, this is a real problem that we have for Startup School coming up. We have 6,000 people or we're going to try and, you know, one of the experiments we're going to try and do is can we seat 800 people at a time perfectly clustered? So, the person sitting to the left and the right of you is the perfect person for you to meet at Startup School. We have to do that in deterministic space combined with latent space. The computation this computation this actual storage like where everyone is inside like, you know, this multi-dimensional array of 800 seats um it actually must not live in the context window. The LLM has to do the human part and seat people. It's actually exactly what you would do if um you know, you were a human tasked with this thing. You would probably have to physically print out 800 pages and go in a a big room and like say like, well, where does this person go? Only now it can all happen in your computer and it can instead of taking a month, it might be able to you might be able to do it with you know, couple hundred dollars worth of tokens and probably 10 minutes.

所以，如果这场演讲你只记住一件事，那就记这个——这也是我一路摸索才发现的：你必须非常非常小心地搞清楚，计算（computation）到底发生在哪里。它几乎总是发生在两个不同的地方。我们碰到的所有 bug、所有 AI 工程上的问题，通常都是因为某件事发生在了等式的一边，而它本该在另一边。第一个地方我称之为潜空间（latent space），也就是 LLM 本身。你用它来做什么？品味、判断力、理解一个人说得含糊时真正想要什么——这些非确定性的调用，是活在模型里的计算，你用 markdown 文件来引导它。而第二个地方是确定性空间（deterministic space），这是工程师们熟悉的领域：你的 code agent 跑出去写 TypeScript，或者如果你用 Elixir，可能在写 Erlang。给大家举个真实的例子，这是我们即将到来的 Startup School 面临的实际问题：我们有 6000 人参加，我们想做一个实验——能不能一次让 800 个人入座，而且座位完美聚类？也就是说，坐在你左边和右边的人，恰好是你在 Startup School 最应该认识的人。这件事必须在确定性空间和潜空间的结合中完成。这个计算、这个实际的存储——每个人在这个 800 个座位的多维数组里的位置——绝对不能放在 context window 里。LLM 只负责做「人」的那部分工作，去安排座位。这其实正是一个人类被派去做这件事时会做的：你大概得把 800 页资料打印出来，跑到一个大房间里，一个个琢磨「这个人该坐哪儿？」只不过现在这一切都可以在你的电脑里完成，原本要花一个月的事，现在可能只需要几百美元的 token 和大概 10 分钟。

[10:44] Garry

And so, that's, you know, I would argue that's pretty remarkable. These are things that you couldn't do even I don't know, 6 months ago. Which brings me to working memory and that's, you know, sort of my favorite way to understand it is you and I human beings we only hold about seven things in our head at once. 7 plus or minus 2. It's one of the most famous papers in cognitive psychology and it's why local phone numbers are seven digits and why you forget the eighth item on your grocery list. Uh that's the entire working memory generally of a human being and every institution humanity has ever built, every checklist, every org chart, every filing cabinet is a prosthetic for that limit. It's kind of a wild thing to think about. But an AI agent holds a million tokens. That's about a thousand pages. I was trying to explain to my 10-year-old what G brain was recently. I said, "There's, you know, the AI agent can keep about three Harry Potter books sitting open in its head all at once. And it can find a needle in any of them and synthesize across all three in seconds. And that's quite magical, actually." Three Harry Potter books versus seven digits. I mean, that's pretty awesome. I mean, I don't know. Is that AGI? Maybe not, but it's already a very different operating and regime. Almost every company on the earth is still running an org that's designed for the seven-digit brain. But notice what that also tells you. Three books is a lot, but it's also very little. Your company is not three books. Your company is a library. Every email, every meeting, every decision, its reasoning, every customer conversation, every postmortem. The question that determines whether your agents are geniuses or goldfish is who decides which three books are open on that desk. That's context engineering.

我认为这相当了不起。这些事情哪怕在 6 个月前你都做不到。这就引出了工作记忆（working memory）这个话题——这是我最喜欢的理解方式：你我这样的人类，脑子里一次只能装大约 7 件事，7 加减 2。这是认知心理学里最著名的论文之一，也是为什么本地电话号码是 7 位数、为什么你总会忘记购物清单上的第 8 项。这基本就是一个人类的全部工作记忆。而人类建立的每一个制度——每一份清单、每一张组织架构图、每一个文件柜——都是为了弥补这个极限的「假肢」。想想还挺震撼的。但一个 AI agent 能装下 100 万个 token，大约相当于 1000 页。我最近想跟我 10 岁的孩子解释 G-brain 是什么，我说：「AI agent 可以同时在脑子里摊开大约三本《哈利·波特》，它能在任何一本里大海捞针，还能在几秒钟内把三本书的内容综合起来。这其实挺神奇的。」三本《哈利·波特》对比 7 位数字——这真的很了不起。这算 AGI 吗？也许不算，但这已经是一个完全不同的运行范式了。地球上几乎所有公司，运转的组织架构仍然是为「7 位数字的大脑」设计的。但注意，这也告诉你另一件事：三本书很多，但同时也很少。你的公司不是三本书，你的公司是一座图书馆——每封邮件、每场会议、每个决策及其推理过程、每次客户对话、每份复盘报告。决定你的 agent 是天才还是金鱼的问题在于：谁来决定桌上摊开的是哪三本书？这就是 context engineering。

[12:53] Garry

And this is what a company brain is. It's the library plus the librarian. Now, some of you are already thinking, "This is just rag." And you're right that retrieval is the primitive, the same way Postgres is just B-trees. The hard part is everything around it. What gets written down in the first place into the knowledge wiki, how it gets enriched and linked, what gets promoted to hot memory versus filed as cold reference, who arbitrates when two facts disagree. Retrieval is easy. Being worth retrieving from is the product. So, I've been building mine in the open. It's called It works with any harness, but it loves open client Hermes agent. It's effectively Postgres for agents, a retrieval layer whose job is to figure out for uh for any three you know, for any task, what three books should be loaded into the agent's head. My personal one started as a rooms full of books or so. Now, it's a warehouse, about 220,000 pages written mostly by my agents from my email, meetings, 20 years of notes, uh and the lived experience of me. And that's the point. It's my second brain. And when a founder emails me about a crisis, before I start reading as before I even finish reading that email, my agent has already pulled every prior conversation with that founder, three portfolio companies that hit the same wall, and what actually worked for those people. Uh when my agent does anything, it does it does everything knowing what I already know. And that's the difference between an assistant and a colleague. So, let me stress test my own pitch because you would anyway. Company brains do have failure modes. Uh a brain nobody curates becomes a garbage dump with great search. Retrieval will surface a stale fact with total confidence. Um a bad skill file encodes a bad process forever. Uh that's bad. So, primitive the primitive is not memory. It's memory plus hygiene, provenance on every fact, contradiction contradiction checks when new information collides with the old, and a librarian, human plus agent, whose actual job is pruning. Treat the brain like a production infrastructure, and it compounds. Treat it like a dumping ground, and you get a very confident agent that is wrong in ways nobody can trace.

这就是「公司大脑」（company brain）的含义：图书馆加上图书管理员。现在，你们有些人已经在想：「这不就是 RAG 吗？」你说得对，检索确实是底层原语，就像 Postgres 本质上「不过是 B 树」一样。难的是围绕它的一切：一开始什么东西该被写进知识 wiki，内容如何被丰富和关联，什么被提升为热记忆、什么被归档为冷参考，当两个事实互相矛盾时由谁来仲裁。检索很容易，「值得被检索」才是产品。所以我一直在公开构建我自己的这套东西，它叫 G-brain。它兼容任何 harness，但它最爱 OpenClaw 的 Hermes agent。它实际上是「给 agent 用的 Postgres」——一个检索层，它的工作就是为任何任务搞清楚：该往 agent 的脑子里装载哪三本书。我个人的这套系统，一开始只有几个房间的书那么大，现在已经是一座仓库了，大约 22 万页，大部分是由我的 agent 写的，来源是我的邮件、会议、20 年的笔记，以及我本人的亲身经历。这就是重点：它是我的第二大脑。当一个创始人发邮件跟我说他遇到了危机，在我还没读完那封邮件之前，我的 agent 已经调出了我和这位创始人之前的每一次对话、三家撞过同一堵墙的 portfolio 公司、以及当时对他们真正起作用的方法。我的 agent 做任何事情时，都带着我已有的全部知识在做。这就是「助理」和「同事」的区别。那么，让我来对自己的这套说辞做一次压力测试——反正你们也会这么做。公司大脑确实有失败模式：一个没人打理的大脑，会变成一个配了顶级搜索的垃圾场；检索会以十足的自信把一个过时的事实端到你面前；一个糟糕的 skill 文件会把一个糟糕的流程永久固化下来——这很糟糕。所以，真正的原语不是记忆，而是「记忆加卫生」（memory plus hygiene）：每条事实都要有出处（provenance），新信息与旧信息冲突时要做矛盾检查，还要有一个图书管理员——人加 agent——他的本职工作就是修剪。把这个大脑当作生产级基础设施来对待，它就会复利增长；把它当垃圾场，你就会得到一个非常自信、但错得没人能追查的 agent。

[15:13] Garry

And um here's the discipline that I think, you know, personally uh makes our company brain and my personal AI compound. Um that's my signature move. And, uh, you know, it's what I say to every YC company and every, uh, everyone inside YC, which is never do one-off work. You can open open claw, you can open your harness, you do some work, but then when you're happy, you know, and it'll come back. It's, you know, kind of a bad job. It's kind of like an intern that's not that good. But the great thing is you can just say, "Hey, I didn't like that. Fix it." Right? I'm sure all of you do this. But don't stop there. You actually need to at the end of that task, uh, skillify it. And so I have a blog post on X about that. You can search for skillify it and, you know, go get that skill file and then just load it into your your own harness and it'll just turn whatever you just did into a skill that you can reuse. Because if you have to ask for something twice, you failed. Um, so yeah, if you remember only one thing, it's that. Like when you, you know, use your AI agent and then when you're done with it and you're happy with the output, skillify it. It's going to be awesome. Um, the organization that captures what it learns like this gets smarter every single day. The one that doesn't wakes up every morning with amnesia, no matter how good the model is. Model quality is rented, but if you build your brain, your you own that brain. So, Theo's question head-on. What do we build now? Build the AI native company, not a company that just uses AI. A company that is shaped like what I just described from day one. A thin team, skill files for everything, the founder still in the code, library, this library, this company brain, your personal AI. Uh, use G brain if you'd like. Uh, it's open source and free, but you don't have to. There are a lot of really good ones. The the library will compound from the first week and your whole org will be wired to run at about 400X.

接下来说说我个人认为让我们的公司大脑和我的个人 AI 得以复利增长的那条纪律。这是我的招牌动作，也是我对每一家 YC 公司、对 YC 内部每一个人都会说的话：永远不要做一次性的工作（never do one-off work）。你可以打开 OpenClaw，打开你的 harness，干点活，结果回来了——可能干得不怎么样，有点像一个不太行的实习生。但妙就妙在你可以直接说：「嘿，我不满意，改一下。」我相信你们都这么干过。但别止步于此。在任务结束时，你需要把它「技能化」（skillify it）。我在 X 上发过一篇关于这个的帖子，你可以搜 skillify it，把那个 skill 文件拿下来，加载进你自己的 harness，它就会把你刚才做的事情变成一个可以复用的技能。因为如果同一件事你需要开口问两次，你就已经失败了。所以，如果你只记住一件事，就记住这个：用完你的 AI agent、对结果满意之后，把它技能化。效果会非常棒。像这样把学到的东西沉淀下来的组织，每一天都在变聪明；而不这么做的组织，无论模型多好，每天早上醒来都患着失忆症。模型的质量是租来的，但如果你构建了自己的大脑，那个大脑就是你自己的。所以，正面回答 Theo 的问题：我们现在该造什么？造 AI 原生的公司，而不是一家「只是在用 AI」的公司。一家从第一天起就按我刚才描述的样子塑形的公司：精简的团队，万事皆有 skill 文件，创始人依然亲自写代码，再加上这座图书馆——这个公司大脑、你的个人 AI。愿意的话可以用 G-brain，它开源免费，但你不必非用它，市面上有很多很好的选择。这座图书馆从第一周起就会开始复利，你的整个组织将以大约 400 倍的效率运转。

[17:15] Garry

And if you want the green field, the thing that I'd build if I were 25 and sitting where you're sitting, every company on this earth is about to need a brain. The memory layer that means that you never have to re-ask what you knew. Personal AI that actually knows you. We're building G brain in the open and MIT open source. Um I'm not trying to make money from this because I think the layer should be open the way Linux is open. But the layer itself, company brains, personal context, the librarian that picks the three books, that's all wide open territory. I hope somebody builds the defining company here. And I'd like to fund you at YC if you do. Now, let me be honest in a way that maybe undercuts my own pitch. You don't need my tools to start. Open Claw is the Ferrari. I will always recommend it, but Codex is a really good Honda. It will do 90% of this. Uh it will not blow your face off, but it will get you there. Use whatever. The concepts are the point, not my repos. You know, think about where the computation is. Use skill files as employees. The librarian the librarian. Never do one-off work. Those travel with you to any stack. So, let me land this. A lot of people in the world right now are terrified about what happens to all the jobs, and I understand the fear. But I want to say it plainly, that is a failure of imagination. And the people in this room are the answer to it. What I just described, you're going to take to your startup. You will multiply yourself, and every person in your company will multiply themselves, and you will go build the companies that become the beacon for how all of this works in society. Abundance is not a policy paper, it is shipped software. I have a friend who has a rare form of epilepsy. He built a repo of 80,000 markdown files, a company brain for one small boy, and he pushed himself to the absolute edge of human what humanity knows about his son's exact condition.

如果你想要一片未开垦的处女地——如果我现在 25 岁、坐在你们的位置上，我会造的东西是：地球上每一家公司都即将需要一个大脑。一个记忆层，让你永远不必重复追问你已经知道的东西；一个真正了解你的个人 AI。我们正在公开构建 G-brain，MIT 开源协议。我不打算靠它赚钱，因为我认为这一层应该像 Linux 一样开放。但这一层本身——公司大脑、个人上下文、那个挑选三本书的图书管理员——是一片完全敞开的疆域。我希望有人能在这里造出定义性的公司，如果你做到了，我很乐意在 YC 投资你。现在，让我说句可能会拆自己台的实话：你不需要我的工具才能开始。OpenClaw 是法拉利，我永远会推荐它，但 Codex 是一辆很不错的本田，它能搞定这里说的 90%。它不会让你惊掉下巴，但能把你送到目的地。用什么都行，重点是这些概念，不是我的仓库：想清楚计算发生在哪里；把 skill 文件当员工用；图书管理员；永远不做一次性工作。这些东西可以跟着你迁移到任何技术栈。那么，让我来收尾。现在世界上很多人对「工作会怎么样」感到恐惧，我理解这种恐惧。但我想直白地说：那是想象力的失败。而这个房间里的人，就是它的答案。我刚才描述的这一切，你们会带回自己的创业公司。你会让自己成倍放大，你公司里的每一个人都会成倍放大自己，然后你们会去创办那些成为灯塔的公司，向整个社会示范这一切该怎么运作。「丰裕」（abundance）不是一份政策白皮书，它是被交付出去的软件。我有一个朋友，他的孩子患有一种罕见的癫痫。他建了一个包含 8 万个 markdown 文件的仓库——一个为一个小男孩而建的公司大脑——他把自己推到了人类对他儿子这种疾病认知的最前沿。

[19:15] Garry

No lab, no grant, no permission. A father, a laptop, and a library. That's not a side story. That is the exact architecture I've been describing for the last 20 minutes. A library of the librarian, the right three books open at the right moment, pointed at the thing this man loves the most in the world. You can do that now. Every problem where you thought, I wish I had that person but I can't get them. You can. Every code base you thought was too buggy to fix, you can fix all of it. Every archive too big to read, every data set too gnarly to clean, every ocean you were told not to boil. We can boil the ocean now.

没有实验室，没有科研经费，没有任何人的许可。一位父亲，一台笔记本电脑，一座图书馆。这不是什么花边故事——这正是我过去 20 分钟一直在描述的那套架构：一座图书馆，一位图书管理员，在正确的时刻摊开正确的三本书，然后对准这个男人在这世界上最爱的那个人。你现在就能做到这些。每一个你曾想过「真希望我能请到那个人，但我请不起」的问题——现在你可以了。每一个你以为 bug 多到没法修的代码库——你可以全部修好。每一座大到读不完的档案库，每一份乱到没法清洗的数据集，每一片别人告诉你「别妄想煮沸」的大海——我们现在可以把大海煮沸了。

[20:05] Garry

[applause] [applause]

（掌声）

[20:11] Garry

And every single one of you can fly. Not metaphorically, mechanically. And you need to to survive, to thrive, to win. Theo asked, "What we should build now?" And here's the whole damn whole answer. Build that AI native company and build it build the thing underneath it. The brain, the memory, the compounding library. That makes every company after yours easier to build. Go boil the ocean. Go write that test. Go ship that skill. Some of the companies you're about to watch in the battlefield are already doing this. Go build the one that does it best. Thank you.

你们每一个人都可以飞起来——不是比喻意义上的，是机械意义上的。而且你必须飞，才能生存、才能繁荣、才能赢。Theo 问：「我们现在该造什么？」这就是完整的答案：去造那家 AI 原生的公司，同时去造它底下的那层东西——大脑、记忆、那座会复利的图书馆——让你之后的每一家公司都更容易被创办。去煮沸大海，去写那个测试，去交付那个 skill。你们接下来将在 battlefield 环节看到的一些公司，已经在这么做了。去创办那家做得最好的公司。谢谢大家。