AI Engineer World's Fair 2026
Every company should have a Brain — Garry Tan, Y Combinator
Read the talk
Every company needs a brain
Skills organize agent work, code preserves operational state, and a curated company memory turns each completed task into capability the organization can reuse.
From a talk by Garry Tan
What do we build now?
What should you build when one person might be able to do work that previously required an entire organization? Garry Tan approaches that question as a founder, investor, and leader transforming Y Combinator, a twenty-year-old institution, into an AI-native company. His ambition is literal: one person doing work that once took a thousand people. With the startup battlefield about an hour away, the immediate challenge for the audience is to recognize what has become possible before deciding what to build.
The first example is his own engineering work. In 2013, Tan was building YC’s internal social network while also investing in companies. He recalls producing about 14 usable logical lines of code per day as a near-full-time engineer, excluding comments. He places that beside an asserted contemporary range of 15–50 lines per day and a median near 15. Today he runs YC full-time, spends fewer hours engineering, and has a 5 p.m. child pickup. Tan reports roughly 400× his earlier raw code output, with an asserted 8× floor and 80× middle after discounting verbosity and scaffolding. These are personal estimates; the talk supplies neither a measurement protocol nor the calculation behind those discounts.
The proposed source of the gain is how the work is organized. Tan contrasts people getting 2× and 100× improvements while using the same Claude weights, context window, and API. He reports that a quarter of YC’s Winter ’25 companies had codebases that were 95% AI-generated, and calls that batch YC’s fastest-growing and most profitable. Separately, he says 94 companies funded by YC at seed have crossed $100 million in revenue. He explicitly does not establish that AI-generated code caused the growth. The operating distinction he draws is between using AI as autocomplete and organizing it as a workforce.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
An organization expressed in files
An agent system already has recognizable management functions. A skill file specifies one capability and one job clearly enough to execute. When CLAUDE.md grows too large, a resolver table can direct the agent to the instructions relevant to the incoming task. Tan’s example is loading tests.md when a task requires altering a test. Instead of putting every procedure into the initial context, the system chooses which procedure to bring into the work.
| Agent mechanism | Organizational function |
|---|---|
| Skill file | Employee with a defined job |
| Resolver table | Org chart routing incoming work |
| Filing rules | Internal process and compliance |
| Trigger evaluation | Performance review of routing behavior |
The routing decision itself needs evaluation. Does a request to change a test actually load the intended test instructions? Tan calls this a performance review because a written responsibility is not enough: the system must demonstrate that it follows it. He says tests.md in the routing example and test.md in the evaluation example; an implementation needs one exact filename shared by the route and its check.
A small TypeScript example makes that contract explicit, using tests.md consistently:
typescript
import assert from "node:assert/strict";
import { readFile } from "node:fs/promises";
type TaskKind = "alter-test";
const instructionsFor: Record<TaskKind, string> = {
"alter-test": "tests.md",
};
async function loadInstructions(
task: TaskKind,
load: (path: string) => Promise<string>,
): Promise<string> {
return load(instructionsFor[task]);
}
async function checkTestRoute(): Promise<void> {
const loaded: string[] = [];
const text = await loadInstructions("alter-test", async (path) => {
loaded.push(path);
return "Test-editing instructions";
});
assert.deepEqual(loaded, ["tests.md"]);
assert.equal(text, "Test-editing instructions");
}
async function main(): Promise<void> {
await checkTestRoute();
const instructions = await loadInstructions("alter-test", (path) =>
readFile(path, "utf8"),
);
console.log(instructions);
}
main().catch((error: unknown) => {
console.error(error);
process.exitCode = 1;
});
This checks loading after a task has been classified as alter-test; a full trigger evaluation would also check whether the agent recognizes a natural-language test-editing request. The organizational machinery can therefore be mostly Markdown, with code where an explicit contract or check is useful. In Tan’s framing, working through Claude Code or Codex becomes hiring, training, and managing that workforce.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Small teams, maintained procedures
Tan points to two companies as examples of the economics he associates with this structure:
| Company | Scale reported by Tan | Team size reported by Tan |
|---|---|---|
| Emergent | Nine figures of ARR within eight months of public launch | 15 people at an earlier $15 million ARR milestone |
| Retell | $60 million; the talk does not specify the revenue period | About 40 people |
Emergent, which Tan identifies with YC’s Summer ’24 batch, also reports the $100 million ARR milestone on its own site. ARR is an annualized recurring-revenue measure, not realized annual revenue. Retell’s own announcement labels its figure $60 million ARR and gives 35 employees; that is a separate company account, not a replacement for Tan’s approximate headcount. His broader assertion that these revenue-per-person levels are unprecedented across software, oil, and railroads is not established by the examples alone.
The organizational mechanism is to encode sales, support, operations, and finance as written procedures that agents execute. Engineers maintain those skills and do the work that the skills cannot yet handle. Tan even offers tax filing as a task that could be represented by a skill. This changes the staffing question from how many people to hire for each function to which procedures can run reliably and where people still need to intervene.
He pictures roughly 400 founders at laptops in YC’s batch room, each doing a former person-year of work in a day. That is his competitive ambition, not a measured throughput result for the room. The extension beyond engineering is more concrete: YC’s media, events, and finance staff are creating skill files and cron jobs, including people who had never opened a terminal. Tan reports that a finance employee consolidated about 100 Excel workbooks into one app using YC’s internal OpenClaw and company brain. Her role becomes managing agents rather than first becoming a programmer. Tan extends his 400× aspiration to the organization as a whole.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Keep judgment in the model and state in code
Reliable agent work depends on deciding where computation belongs. Tan separates two places:
- Latent space: The model handles taste, judgment, and interpreting what someone means by a vague request. Markdown instructions steer these nondeterministic calls.
- Deterministic space: Ordinary software performs explicit operations and maintains state. Agents can write that software in TypeScript or other languages; Tan also names Erlang and Elixir.
A recurring source of bugs, in his experience, is putting work on the wrong side of this boundary. Better instructions cannot substitute for an explicit representation of the state a system must preserve.
Startup School seating makes the boundary tangible. Tan describes a prospective experiment involving a population of 6,000 people, seating 800 at a time so that each attendee’s neighbors are especially useful people to meet. The model can judge interpersonal fit, but the multidimensional arrangement of seats must live outside its context window. The seating plan is operational state; interpreting why two people should meet is judgment.
A human given this assignment might print 800 pages, spread them through a room, and move people into groups. The room and paper would provide external memory while the person made the matching decisions. The software version needs the same division of responsibility. Tan speculates that the seating task could take a couple hundred dollars in tokens and probably ten minutes instead of a month. This is a proposed experiment, not a completed benchmark; he presents it as an example of work he believes was out of reach even roughly six months earlier.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Three open books inside a much larger library
External memory is already central to human organizations. Tan invokes George A. Miller’s The Magical Number Seven, Plus or Minus Two, using seven-digit phone numbers and a forgotten eighth grocery item to illustrate limited working memory. Miller’s paper concerns task-dependent limits, learned chunks, and recoding; it does not establish a universal seven-digit ceiling or the origin of telephone-number lengths. The useful organizational observation is that checklists, org charts, and filing cabinets let people coordinate work they cannot keep in their heads.
Tan compares a million-token agent context to about 1,000 pages or three Harry Potter books open at once. Explaining GBrain to his ten-year-old, he describes an agent finding information and synthesizing across those books in seconds. The page and book counts are illustrative, and no model-specific retrieval evaluation accompanies them. He does not treat this as proof of AGI; it is a different working-memory regime from the one around which familiar organizations developed.
But even three open books are small beside a company’s library. That library includes emails, meetings, decisions and their reasoning, customer conversations, and postmortems. Context engineering decides which books are open for the current task. A large context window supplies working space; it does not itself decide which parts of the organization’s history deserve that space.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
The library needs a librarian
The company brain is the library plus the librarian. Retrieval-augmented generation supplies a primitive, but calling the whole system RAG leaves out much of the product—much as describing Postgres as B-trees leaves out the surrounding database system. The harder responsibilities are deciding what enters the knowledge wiki, enriching and linking it, promoting useful material to hot memory, filing other material as cold reference, and arbitrating when facts disagree.
“Being worth retrieving from is the product.” Tan calls his implementation GBrain, describing it as Postgres for agents: a retrieval layer that chooses the relevant books for a task. He says it works with any harness and particularly likes OpenClaw and Hermes Agent. The architectural responsibility remains context selection, whichever harness performs the work.
Tan reports that his personal GBrain archive contains about 220,000 pages, mostly written by agents from email, meetings, and twenty years of notes. When a founder emails about a crisis, he says his agent retrieves prior conversations with that founder, examples from three portfolio companies that encountered similar problems, and what worked for them—before he finishes reading the message. The benefit is not merely finding a document. It is bringing relevant experience into the current decision so the agent can work with knowledge Tan already has. That shared context is what makes it feel more like a colleague than an assistant.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Memory is production infrastructure
A company brain also preserves mistakes. Without curation, it becomes a garbage dump with excellent search: an agent can retrieve a stale fact confidently, while a bad skill file keeps reproducing a bad process. More memory does not fix either failure.
Memory needs hygiene as part of its operating model. Tan identifies three responsibilities:
- Provenance: Preserve the source of every fact so a claim can be traced.
- Contradiction checks: Detect when incoming information conflicts with existing knowledge and resolve the collision.
- Pruning: Give a human-plus-agent librarian explicit responsibility for maintaining the collection.
Treating the brain as production infrastructure makes accumulated knowledge useful over time. Treating it as a dumping ground produces confident errors that nobody can reconstruct.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Capture the procedure after correcting the work
The complementary discipline is to avoid one-off work. An agent’s first attempt may be disappointing, and asking it to fix the result is only the beginning. The learning loop should continue until the corrected procedure becomes reusable:
- Give the agent a task in your chosen harness.
- Inspect its output and correct what is unsatisfactory.
- Once the result is good, turn the completed workflow into a skill.
- Reuse that skill instead of teaching the same procedure again.
Tan calls the capture step “skillify it” and points listeners to an X post and a skill file for doing it in their own harness. His deliberately strict standard is that asking for the same thing twice means the learning was not captured the first time.
The timing matters: capture the process after the output is satisfactory, so the reusable skill includes the corrections that made the work good. An organization that does this accumulates capability each day. Without it, each new session can begin with the same organizational amnesia, regardless of model quality. “Model quality is rented,” as Tan puts it; the brain built from the organization’s experience is something it owns.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Build the company and the memory underneath it
This produces a concrete company blueprint: a thin team, skills for repeatable work, a founder who remains in the code, a company library, and personal AI that can use it. Start capturing knowledge early enough that the library begins compounding in the first week. GBrain is offered as a free, MIT-licensed option, with alternatives explicitly welcome. The goal is an organization shaped around these capabilities from the beginning.
The infrastructure itself is also a product opportunity. Company brains, personal context, and the librarian that selects the right information remain territory to build in. Tan wants the underlying memory layer to be open in the way Linux is open, rather than making GBrain itself his commercial business. He invites founders to build a defining company around this need—and offers YC as a potential funder.
Adoption does not depend on choosing his repositories. Tan likens OpenClaw to a Ferrari and Codex to a dependable Honda. He estimates that Codex can do 90% of this, without defining an evaluation behind the percentage. The portable parts are the placement of computation, skills with clear responsibilities, a library with a librarian, and the habit of capturing completed work. Those principles can travel across stacks.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
A library pointed at a problem that matters
The ending moves from company design to the fear of job displacement. Tan acknowledges that fear, then argues that founders who multiply their teams’ capabilities can demonstrate a more abundant alternative. His formulation is direct: “Abundance is not a policy paper, it is shipped software.”
His closing example is a father using a research library to investigate his son’s rare epilepsy condition. Tan describes a repository of 80,000 Markdown files assembled for that research. Without a lab or grant, the father uses a laptop and a library to explore the limits of available knowledge about the condition. The architecture is the same one described for a company: a collection, a librarian, and the relevant material brought into context at the moment it is needed. The anecdote concerns research reach; it does not establish a treatment or medical outcome.
The same pattern expands the set of projects worth attempting: work that once seemed to require inaccessible expertise, codebases too buggy to repair, archives too large to read, and datasets too difficult to clean. Tan turns the warning against boiling the ocean into permission to attempt larger problems, describing the audience’s ability to fly as mechanical rather than metaphorical. The concrete challenge is to build both the AI-native company and the brain underneath it—the memory and compounding library that make the next company easier to build. He makes the challenge immediate—write the test and ship the skill—then returns to the battlefield companies already trying and urges the audience to build the one that does it best.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Resources
From the talk
Open-source agent memory system with markdown knowledge storage, retrieval, reusable skills, routing, and evaluation tools.
Personal AI assistant project with public source code and setup instructions.
Nous Research's agent project, named as a harness for GBrain.
George A. Miller's 1956 paper on information processing, immediate memory, chunking, and recoding.
Read the complete timestamped transcript
- 0:00
[upbeat music] Okay, great. Hey, everyone.
- 0:15
How's everyone doing? [cheering] All right. Are we ready for the revolution? Okay. Theo just asked the right question: what do we build now?
- 0:27
I'm gonna answer it from the other side of the table. Uh, I'm a founder, I'm an investor, and I run a twenty-year-old institution that is becoming AI native right now, which is a strange and wonderful thing to do to a twenty-year-old institution.
- 0:43
Uh, and I'll spend about twenty 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
- 1:07
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.
- 1:25
So let me start with a number and, uh, you know, I got torn apart on the internet for this, but I'm gonna say it again in front of all of you anyway.
- 1:32
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 twenty thirteen, I was a YC partner, uh, building the internal social network, um, at, at YC.
- 1:44
Uh, I was also investing in companies but, you know, I was also a near full-time engineer. Um, and when I was doing that, I could maybe do about fourteen usable logical lines of code a day.
- 1:56
Take out the comments, take out all the bullshit, 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.
- 2:07
Like, some people write fifteen, some people write fifty. 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.
- 2:16
Um, that's about median fifteen. 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.
- 2:28
Uh, you know, but I have a five PM kid pickup now, and I did the math on my output, and it's about four hundred X. Now, before the skeptic in the third row right there deflates the number fro-for me, let me deflate it myself.
- 2:41
If you don't trust the raw code, well, fine. Take the most pathological verbosity pla- penalty you can stomach and assume the agent writes bloated code, assume half of it is scaffolding, assume I'm flattering myself.
- 2:53
It's still eight X at the floor and eighty X 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.
- 3:05
It's not the model. The two X people and the hundred X people are using the exact same Claude, same weights, same context window, same API.
- 3:17
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 ninety-five percent AI-generated, and that was a year ago.
- 3:32
That batch has become the fastest-growing, most profitable batch in the history of YC. Ninety-four companies total have now crossed a hundred million dollars in revenue from a seed check in the history of YC.
- 3:45
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 autocomplete.
- 3:58
They're treating it as a workforce. The companies that wired the work differently are the ones that are bending the curve. [smacks lips] So what does wiring the work really mean? No slides.
- 4:11
This is the heart of the talk, and this is what I most want you to steal. Everything we've learned building with agents maps to an organization.
- 4:21
There's no slides.
- 4:23
No slides, no slides.
- 4:23
Sorry, there's no slides. [laughs] I have no slides. I'm so sorry. [cheering]
- 4:31
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 Claude Code and it says your context is too big in claude.md, you run off and create a resolver table.
- 4:52
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 tests.md, and you have a whole table of these things.
- 5:02
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-- is actually in compliance and, uh, trigger evals.
- 5:23
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?
- 5:36
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.
- 5:52
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 Code or Codex, you're not writing software, you're hiring, training, and managing a workforce made of markdown.
- 6:10
Uh, and, you know, that's-- there are tons of companies at YC 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.
- 6:22
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.
- 6:38
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.
- 6:52
The AI-native companies that 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.
- 7:07
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 [REDACTED:location]. Like, it actually kinda looks like this.
- 7:20
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.
- 7:35
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.
- 7:43
Here's the extension, uh, most engineering talks miss. It's not just the engineers. Um, at YC, uh, as we make our transformation, it's our media people, our event staff, our finance team.
- 7:54
People who've never opened a terminal in their lives are building skill files and cron jobs, and, uh, one of our finance folks just collapsed about a hundred Excel workbooks into a single app she built with, uh, our internal OpenClaw and Company brain.
- 8:11
She's not a programmer. She's a manager of agents now, and everyone at YC now is.
- 8:17
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.
- 8:29
It's not just 400x engineers. It's one company that operates at the level of 400x everyone else. And so, you know, if you remember, uh, only one thing about this, I mean, this is one of the things I had to discover along the way.
- 8:43
Like, you actually have to be really, really careful about where the compute-- computation is actually happening. Uh, 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, uh, something is happening in one side of the equation that should be in
- 9:02
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.
- 9:14
The non-deterministic calls, the computation that lives in the model, and you steer it with the markdown file. Uh, and then deterministic space is what, uh, engineers know. Like your, your code agents go off and write TypeScript or, you know, maybe they're writing, um, you know, Erlang if you're using Elixir.
- 9:33
Um-
- 9:34
Whoo.
- 9:34
Yeah. Deterministic space is the second place. Um, let's say, you know, this is a real problem that we have for startup school coming up. We have six thousand people, or, uh, we're gonna try and...
- 9:44
You know, one of the experiments we're gonna try and do is can we seat eight hundred people at a time, um, 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.
- 9:56
Um, we have to do that in deterministic space combined with latent space. Uh, the computation, this computation, this actual storage of, like, where everyone is inside, like, you know, this multidimensional array of eight hundred seats, um, it actually must not live in the context window.
- 10:15
The LLM has to do the human part and seat people. Um, it's actually exactly what you would do if, um, you know, you were a human tasked with this thing.
- 10:25
You would probably have to physically print out eight hundred pages and go into a big room and, like, say, like, "Well, where does this person go?" Only now it can all happen, um, in your computer, and it can-- instead of taking a month, it might be able to-- you might be able to do it with, uh, you
- 10:39
know, a couple hundred dollars' worth of tokens and probably ten minutes. Um, and so that's, you know-- I would argue that's pretty remarkable. These are things that you couldn't do even, uh, I don't know, six months ago.
- 10:53
Um, which brings me to working memory and, uh, that's, you know, sort of my favorite way to understand it is, um, you and I, human beings, we only hold about seven things in our head at once.
- 11:05
Uh, "Seven Plus or Minus Two," it's one of the most famous papers in cognitive psychology, and it's why, uh, local phone numbers are seven digits and why you forget the eighth item on your grocery list.
- 11:18
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, which is kind of a wild thing to think about.
- 11:35
But an AI agent holds a million tokens. That's about a thousand pages. I was trying to explain to my [REDACTED:age] what GBrain 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
- 11:55
across all three in seconds," and that's quite magical actually. Three Harry Potter books versus seven digits.
- 12:05
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.
- 12:23
But notice what that also tells you. Three books is a lot, but it's also very little. Your company is not three books. Ev- your company is a library. Every email, every meeting, every decision, its reasoning, every customer conversation, every postmortem.
- 12:39
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, and this is what a company brain is.
- 12:55
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.
- 13:06
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.
- 13:21
Retrieval is easy. Being worth retrieving from is the product. So I've been building mine in the open. It's called GBrain. It works with any harness, but it loves OpenClaw and Hermes Agent.
- 13:33
It's effectively Postgres for agents, a retrieval layer whose job is to figure out for, uh, for any thri-- you know, for any task, what three books should be loaded into the agent's head.
- 13:44
My personal one started as a rooms full of books or so. Now it's a warehouse, about two hundred and twenty thousand pages, written mostly by my agents from my email, meetings, twenty years of notes, uh, and the lived experience of me, and that's the point.
- 13:59
It's my second brain, and when a founder emails me about a crisis, before I start reading this-- before I even finish reading that email, my agent has already pulled every prior conversation with that founder, three portfolio companies, they hit the same wall, and what actually worked for those people.
- 14:15
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.
- 14:27
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.
- 14:46
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.
- 15:03
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.
- 15:13
And, um, here's the discipline that I think, you know, personally, uh, makes m- 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.
- 15:31
You can open OpenClaw, you can open your harness, you do some work, but then when you're hap-- you know, and it'll come back. It's, you know, kind of a bad job.
- 15:38
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.
- 15:46
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
- 16:05
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 y- you know, use your AI agent and then when you're done with it and you're happy with the output, skillify it.
- 16:22
It's gonna 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.
- 16:35
Model quality is rented, but if you build your brain, your, you own that brain.
- 16:41
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.
- 16:52
A thin team. Skill files for everything. The founder still in the code. Library. This library, this company brain, your personal AI. Uh, use GBrain if you'd like. Uh, it's open source and free, but you don't have to.
- 17:04
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 four hundred X.
- 17:16
And if you want the greenfield, the thing that I'd build if I were [REDACTED:age] and sitting where you're sitting, every company on this earth is about to need a brain.
- 17:23
The memory layer that means that you never have to re-ask what you knew. Personal AI that actually knows you. We're building GBrain in the open and MIT open source.
- 17:32
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.
- 17:45
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.
- 17:56
You don't need my tools to start. OpenClaw is the Ferrari. I will always recommend it, but Codex is a really good Honda. It will do ninety percent of this.
- 18:06
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.
- 18:15
Use skill files as employees. The library and 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 wanna say it plainly, that is a failure of
- 18:35
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
- 18:52
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 [REDACTED:gender], and he pushed himself to the absolute edge of humani- what humanity knows about his son's
- 19:14
exact condition. 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.
- 19:28
A library, a l- the librarian, the right three books open at the right moment pointed at the thing this [REDACTED:gender] loves the most in the world. You can do that now.
- 19:38
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.
- 19:49
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. [audience applauding]
- 20:11
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 an- whole answer.
- 20:26
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.
- 20:39
Go write that test. Go ship that skill.
- 20:42
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. [audience applauding] [upbeat music]