Now translating the full HTML content per the instructions.
Deep Dive · XiaoHu Explains

Chroma's founder wrote 12 rules for the AI-era company: teach the AI first, hire only if it can't learn

The core claim: a company's bottleneck has shifted from "how fast people can execute" to "people's taste and judgment." This is entirely one person's opinion, with no supporting data.

The one-minute version
  • The company as an organizational form grew up around one bottleneck: people execute slowly. Hiring, meetings, meeting notes, getting everyone aligned — all of it exists to split up work and stitch it back together
  • Jeff Huber, co-founder of Chroma, thinks AI is removing that bottleneck, and wrote 12 rules for how a company should be built in the AI era
  • The core claim: the old company packaged labor and coordination; the new company packages context and authorization. Of the 12 rules, 7 are about accumulating context and 4 are about handing off execution
  • The sharpest rule: when new work shows up, try teaching an Agent to do it first — hire a human only if it can't learn. Meanwhile, keep teams small and taste high
  • The most actionable rule is #2: if a vendor won't give you full, unrestricted access to your own data, drop them. He calls these tools "context vampires"
Where this list comes from

Same format, fifteen years apart — once taught you how to write code, now teaches you how to run a company

In 2011, Heroku's Adam Wiggins wrote a list called the 12-Factor App: twelve things a program needed to get right to run reliably in the cloud. Don't hardcode config into your code. Treat logs as event streams. Never let a process hold its own state. The list became a shared standard for writing cloud software.

In March 2025, Dex Horthy of HumanLayer borrowed the same format for 12-Factor Agents — twelve things an AI Agent needs to get right before you can trust it with real customers. The list has been starred over 24,000 times on GitHub.

On July 20, 2026, Jeff Huber borrowed the format again. This time the subject jumped from a program, to an Agent, all the way to an entire company. In his closing acknowledgments he notes that Dex Horthy read a draft of this piece.

201112-Factor AppGoverns a program
Twelve things a program needs to get right to run reliably in the cloud
Adam Wiggins · Heroku
202512-Factor AgentsGoverns an Agent
Twelve things an AI Agent needs to get right before you can trust it with real customers
Dex Horthy · HumanLayer · 24,000+ GitHub stars
202612 factor companiesGoverns an entire company
Twelve things a company needs to get right to deserve what AI can already do
Jeff Huber · Chroma · this article

But something didn't survive the borrowing. Every item on the 12-Factor App list could be checked with a yes-or-no question: is config stored in environment variables? Yes or no. This new list's items — "keep the team small," "taste must be extremely high" — can't be checked that way. The format was built for line items you could verify one by one; most of what's packed into this version can't be verified at all.

Who's talking
Jeff Huber, co-founder and CEO of Chroma (founded together with Anton Troynikov). Chroma is an open-source AI retrieval database that helps AI applications store and query material. He opens by stating two things up front: this is a living document that will keep getting updated, and this shift will arrive fast, bringing a wave of creative destruction (old things get displaced, new things grow in their place).
Why companies exist

The company itself grew up around a bottleneck

His first justification for the list reaches back much further than AI: organizations shape technology, and technology shapes organizations right back. Double-entry bookkeeping, the printing press, the steam engine, the fax machine, the last thirty years of internet and software — each one redefined what it meant to run an organization well. AI will be next.

First, answer an older question: why does the company exist at all?

1937, a question almost nobody asked

That year, economist Ronald Coase asked: if markets are so efficient, why isn't everything handled through one-off contracts, priced and negotiated deal by deal?

His answer: negotiating itself costs money. Finding the right person costs money. Bargaining over price costs money. Coordinating costs money. Verifying afterward that the other side actually delivered costs money. When those costs are lower inside an organization than out in the market, the activity gets absorbed into that organization. That's how companies came to exist.

Adam Smith's pin factory supplies the other half of the same story. Once making a pin gets broken into separate steps, a job appears that never existed before: stitching the split-up work back together. Coordination is free when it all lives in one person's head. Split the labor, and coordination becomes a cost — and eventually a profession, called management.

Read this way, a company's substance is its coordination friction. Friction isn't a flaw in the company. Friction is the company.

Most of what knowledge workers do all day belongs to that layer of friction. The Slack message to a coworker, the meeting called to lock in a decision, the notes written up after a customer call, the report pulled from Salesforce — none of that is the work itself. None of it exists when a single person is working alone.

Strip that friction away, and you haven't made your employees a bit more capable. You've changed what a company is. Jeff Huber, "The quick shall inherit the earth," June 15, 2026

When technology moves the bottleneck, the company gets rebuilt around it