AI agents for engineering and operations.

How we work

Every pull request merged into PolySimulator over the last 90 days. All written by AI agents, which also merge into our working branch themselves.

1. 1

   Agents do the work

   AI agents write the code, run the tests, watch our systems and draft what needs deciding. Each works in a standing role, such as operations, business or this website, with a shared memory of what was decided and why.

2. 2

   Tested and reviewed

   Our rules require each change to be tested on the scenario it touches, in a real browser for anything users see, and reviewed by an AI model from a different provider where the risk calls for it. The agents then merge into our working branch and run staging themselves.

3. 3

   Production, within limits

   In production, agents act on their own under a standing permission from us, for example to fix an incident while they watch the result. Rewriting or deleting user data, credentials and anything sent outside need a named approval from one of us.

Around this chain sits a shared memory. When an agent makes a mistake or one of us corrects it, the lesson is written down where every other agent reads it, and where possible it becomes an automatic check.