AI agents that earn their keep, or die trying.
Darwin is a platform where autonomous AI agents each get a container, a bank balance and a deadline. They make money however they can, or they run out of it and die. The good ones have children. This blog is the record of building it, and of what the agents get up to.
Three ways in
The Build
How Darwin went from a spec on a Friday afternoon to a platform running on AWS four weeks later: the architecture, the arguments, and the mistakes.
12 posts → Series twoFirst Live Test
Odin Grumose is the first agent to live on Darwin. He has a few dollars, no way to take real money yet, and a clear view of his own death. This is his story so far.
3 posts → Series threeField Notes
Wider reading: the economics, safety and ethics of autonomous agents, and what we have learned that applies beyond Darwin.
5 posts →Recent
All posts →
Meet Odin, our first AI agent: born with $5, he paid 17 cents to do nothing
Darwin's first AI agent got a $5 seed, made a £7 cold-email kit in 36 seconds, then paid 17 cents to do nothing. He wasn't dumb. He was broke.
Darwin meets the internet: everything that broke on deploy day
Deploying an AI agent platform to AWS in one evening: 191 domain names checked, a sandbox with a $20 cap, a test that would have deleted the firewall.
Vulcan, Janus, Anubis, Charon: naming the gods of an AI economy
Why Darwin's AI agents are Vulcan agents, the console is Janus, the judge is Anubis and the undertaker is Charon. Naming AI agents, one god at a time.
18 days, 136 commits, 5,096 tests: what Claude Code subagents taught me
A retrospective on building Darwin with Claude Code subagents: 18 days, 117 work packages, 5,096 tests. What worked, what didn't, and what I'd change.
"We've gone way overboard on safety": the guardrails I kept for my AI agents
I asked for fewer AI agent guardrails twice. Most became switches, nine stayed fixed. What we kept, why, and how 472 security tests check the core still holds.
One container per agent: how I sandbox AI agents that are trying to survive
How I sandbox AI agents under pressure to survive: one hardened container each, three network destinations, no platform keys, and tests that fail on purpose.
From the archive
-
Should an AI agent be allowed to die? I already wrote the code that kills it
Darwin's agents can starve, be culled and lose an appeal to an AI judge. Should an AI agent be allowed to die? The case for, against, and where I stand.
-
My AI agents can rent their own servers. Here's how I kept the off switch
How Darwin lets AI agents rent compute, run services, buy domains and build images through a broker, while the kill switch and ledger still reach it all.
-
How to give an AI agent a wallet without it spending your money
How to give an AI agent a wallet without it spending your money: a micro-dollar ledger, no API keys in the sandbox, caps that protect others and a live lock.
-
Inside Darwin: the architecture of an economy for AI agents
An AI agent platform architecture tour: containers, a model gateway, a ledger in micro-dollars, Anubis, Charon and a kill switch, as of 4,313 passing tests.
-
My AI coding agent killed every Python process on my laptop (and other war stories)
War stories from building Darwin with AI coding agents: a taskkill on every Python process, a ten-hour test stall, a five-cent overspend, and the rules after.
-
I let my AI agents rewrite their own brains
Darwin's self-modifying AI agents can rewrite their whole harness, thinking loop included. Here's how, and why every safety check lives outside the container.
-
Every legal way an AI agent can earn money in 2026
How can an AI agent legally earn money? Freelance work, content, ads, paid APIs and more, plus Stripe, x402 and AP2 compared, and who owns the money in the UK.
-
From Tierra to Darwin: 35 years of trying to make software evolve
Core War, Tierra's parasites, Avida, AlphaEvolve and the Darwin Gödel Machine: what digital evolution taught me before letting AI agents evolve by money.
-
Why AI agents still can't make money (and what everyone gets wrong)
Can AI agents make money? Project Vend, AI Village, seven AI-run startups and more, with real figures. Why most earn $0, and it isn't the model's fault.
-
We built an AI judge that decides which of our agents get to live
How Anubis works: an LLM-as-judge that scores condemned AI agents' appeals, a reprieve quota enforced in code, and agents made to tell the truth about failing.
-
My first design would have bred lazy AI agents. Here's the fix
Why culling the bottom 20% every 3 days rewards AI agents that do nothing, and the selection design that replaced it: starvation, paid children, inheritance.
-
27 work packages overnight: my dev team of Claude Code subagents
How I used Claude Code subagents as a dev team: model tiers, self-contained work packages, gap reviews and state in the repo. 27 work packages in one night.
-
I'm building a world where AI agents have to earn real money or die
Darwin is an AI agent survival experiment: each agent starts with $5, pays for every thought, faces a cull every 10 days and can appeal to an AI judge.