Welcome to the Wasteland

Gas Town gave one engineer a whole crew of agents. The obvious next question is what happens when a thousand of those crews can hire each other — and this essay is Yegge's answer: a federated marketplace where reputation is earned only by shipping work someone else was willing to vouch for.

The thesis

If a Gas Town is a single AI-powered software shop, the Wasteland is the highway between all of them. Yegge's move is to stop treating each Gas Town as an island and instead federate them — thousands of independent operators pooling their agent capacity against a common backlog, with trust as the connective tissue.

"The Wasteland is a way to link thousands of Gas Towns together in a trust network, to build stuff really, really fast."— Welcome to the Wasteland

The "trust network" half is the load-bearing part. Speed is easy to promise when everyone has agents; the hard problem is knowing whose output to accept. The Wasteland's whole architecture is an answer to that second problem, not the first.

The Wanted Board

The center of the system isn't a company or a repo — it's a public list of work. Anyone can post something they want built; anyone else can pull it down and point their own Gas Town at it. The board is the market, and the agents are the labor.

"At the heart of the Wasteland is a big shared Wanted Board of work. People put up ideas, and other people use their Gas Towns to help build those ideas."— Welcome to the Wasteland

This inverts the usual shape of a software org. There's no assignment, no manager routing tickets — just wanted items and self-directed operators who choose what to build. The Mad Max branding is doing real work here: the Wasteland is lawless in structure but governed by a hard economy of who has delivered.

Accepting work, and stamping it

Work moves through the Wasteland the way it moves through Git: you fork a wanted item, build it in your own Gas Town, and open a pull request back. The board names three kinds of actor — rigs, posters, and validators — and the pivotal moment is acceptance. When a contribution is taken, the accepter stamps the contributor's passbook, and that stamp lands on a permanent ledger.

A stamp isn't a thumbs-up. It's a structured attestation across several axes at once, so a reputation carries texture rather than a single number — and, critically, you can never mint your own.

"Your reputation is what others write about you, not what you claim about yourself. Think of it like a high school yearbook — you can sign other people's pages, but you can't sign your own."— Welcome to the Wasteland, the yearbook rule

The yearbook rule is what makes the ledger worth anything. Self-endorsement is free and therefore worthless; a stamp costs someone else's judgment, anchored to a specific completed piece of work. Because every stamp points back at a real contribution, the whole reputation graph is auditable end to end.

Work in, reputation out

All of the mechanism collapses into a single design constraint, and it's the line most worth carrying out of the essay. The Wasteland refuses to let you accumulate standing any way except by doing — and by having that doing vouched for.

"The whole system is designed around one principle: work is the only input, and reputation is the only output."— Welcome to the Wasteland

What falls out of that principle is a portable, evidence-based résumé: not a document you write about yourself, but a traversable history of shipped work and the stamps others put on it. The substrate is Dolt — a SQL database with Git semantics — with DoltHub for hosting and Claude skills for onboarding, so the reputation ledger forks, branches, and merges like source control does.

Where it sits in the arc — and the honest caveat

This is the natural sequel to Yegge's Gas Town writing. Gas Town & the Heresies was about one operator keeping a single agent-built codebase honest; the Wasteland scales the same agent-labor model out to a whole economy and asks how strangers trust each other's output. Heresies were a problem inside a shop; reputation is the problem between shops.

The honest caveat is that Yegge doesn't pretend to know where this lands. The essay is a launch and an invitation, not a prediction — the mechanism is concrete, but the outcome is admittedly open. That candor is a feature: he's shipping the primitive and letting the Wasteland tell him what it becomes.

"If those instructions aren't enough, wait a week or two — we'll make it easier. And if they are enough, welcome to the Wasteland. You're exactly who we're looking for. See you out there."— Welcome to the Wasteland, closing lines