The Company Brain

Your AI is only as good as the brain you give it.

The judgment that runs your company lives in senior people's heads and fifteen-year-old code. I extract it into a governed, git-owned brain that your team and your AI agents can actually run on. Fixed price. Your infrastructure. Yours to keep.

Book a fit call See the offers

The model was never the bottleneck.
Your context is.

The judgment isn't written down

How you price, what a good deal looks like, why the architecture is the way it is. It lives in three people's heads and a codebase nobody dares refactor. Every AI tool you buy starts from zero on the things that make you money.

Agents guess

Point an agent at your business without a substrate it can trust and it improvises. Confidently. The output looks right, reads well, and is wrong in the specific ways only your senior people would catch. So they re-check everything, and the leverage evaporates.

No one verifies the verifiers

Teams now use AI to review AI. Gates that decide what ships, with no evidence they catch what matters. Ask what would happen if a checker silently degraded, and you get silence. This is the gap that widens as models get stronger, not narrower.

Start with the audit. Own the brain. Keep it governed.

Step 1 · Diagnose
Brain Audit
$5,000 fixed · results in one week
A 42-criterion diagnostic across two axes: how mature your AI build system is, and how mature your knowledge substrate is. Proven on four production systems at a mid-size B2B company.
  • 2 to 4 systems scored, read-only access, every verdict traced to a file
  • Written report with your position mapped and gaps ranked
  • Governance read-out: who verifies your verifiers today
  • 60-minute readout call with prioritized next moves
Step 2 · Build
Brain Sprint
$20,000 · 6 to 8 weeks
A working, governed brain on your infrastructure: one domain's judgment extracted from your people and code into a structure agents retrieve from precisely.
  • Extraction sessions with your domain owners, on your repos
  • Git-owned substrate: plain files, wiki-linked, provenance on every claim
  • Governance ritual installed: freshness flags, curation cadence, review gates
  • Your team and agents trained on it before I leave
Step 3 · Keep it honest
Governance Retainer
Subscription · priced at proposal
Brains rot without maintenance. The retainer is the "verify your verifiers" subscription: I audit the brain quarterly and report drift monthly, so it stays trustworthy as your business moves.
  • Quarterly brain audit against the full rubric
  • Monthly drift report: staleness, broken provenance, gate health
  • Spec and review-gate changes checked before they ship
  • Monthly rolling, cancel any time

Extract. Structure. Govern. Adopt.

01
Extract

Structured sessions with the people who hold the judgment, plus a read of the code and documents where it hides. The goal is the reasoning, not just the facts.

02
Structure

Everything lands in plain, git-owned files: a navigable hierarchy with wiki-links and provenance on every claim, so retrieval is precise instead of probabilistic.

03
Govern

Written specs the agents read at runtime, freshness flags, human review gates, and a weekly curation ritual. Governance here is a discipline, not a promise of autonomy.

04
Adopt

Your team learns to feed it, query it, and challenge it. Your agents get wired to it. Adoption is where brains live or die, so it is part of the scope, not an afterthought.

The platforms will rent you a knowledge layer.
This one is yours.

Microsoft, Google, and Glean are all building horizontal knowledge graphs over your data. They will be good. They will also live inside someone else's walls, priced per seat, shaped by someone else's roadmap, and impossible to take with you.

A brain you own is different: plain markdown in a git repository you control. Any model can read it today. Any model can read it in five years. If I disappear tomorrow, you lose nothing.

  • Portable. Works with Claude today, whatever wins tomorrow. No vendor in the load-bearing path.
  • Inspectable. Every claim carries its source and its freshness. You can audit what your agents believe.
  • Private. Your judgment never leaves infrastructure you control. For regulated and IP-sensitive businesses this is the whole game.
  • Compounding. Every decision, every extraction session, every correction makes it more valuable. It is an asset, not a subscription.

I don't sell anything I don't run.

Live client

Currently advising a European industrial-automation group on exactly this build (under NDA). Their internal teams independently converged on this architecture after testing it against embeddings-based retrieval on their own data.

4 systems scored

The audit rubric is not theory. It has been run end-to-end on four production systems at a mid-size B2B company, every verdict traced to a file, and it found the same two gaps in all four: no distillation, and no verified verifiers.

Dogfooded daily

My own company runs on the same system: hundreds of decision records, specs, and knowledge files that AI agents read, act on, and update every day, with human gates on what matters. This page was drafted through it.

Built for companies with real judgment to protect.

This is for you if
  • You are a B2B company (roughly 100 to 5,000 people) whose edge is accumulated domain expertise
  • Your engineers are already using AI and you can feel the ceiling: good output, wrong specifics
  • You are in a regulated or IP-sensitive business and platform knowledge layers make your legal team nervous
  • You want to own the asset, not rent access to it
This is not for you if
  • You are pre-product: there is not enough accumulated judgment to extract yet
  • You want a chatbot over your documents. That is a feature, not a brain, and cheaper elsewhere
  • You expect autonomy without governance. I install rituals and gates, not magic
  • Nobody senior will give the extraction sessions time. The brain is built from your people; no input, no brain

Antony Evans

I built this for myself first. My own company runs on an AI-native operating model: a git-owned brain holding every decision, constraint, and piece of accumulated judgment, with agents that draft, check, and ship real work against it every day, governed by written specs and verified review gates.

Then a client asked for the same thing, and it turned out the method transfers: extract the judgment, structure it so retrieval is precise, govern it so it stays true. That is what I now do for companies that want to own their brain instead of renting one.

Engagements run through Xanadu Labs LLC. I take a small number of clients at a time, because extraction is senior work and I do it personally.

Start here

Start with the audit.

One week, $5,000, and you will know exactly where your AI build and your knowledge substrate actually stand, and what to do about it. If a Sprint makes sense after that, you will know precisely what it should build first.

Book a fit call or email directly: [email protected]