Varoon Kodithala(@varoonkodithala):at one point, we had a Slack channel dedicated to collecting feedback + improving our agents' model configs, skills, and prompts along the axes we care about. so we decided to embed this into our product! every factory @warpdotdev comes built-in with a functional self-improvement loop: > a default set of scorers that judge agent runs on things like code quality and verbosity > over time, scorer results bubble up into a set of concrete improvements to agents' base config waitlist is open:

2026.08.18 22:33

at one point, we had a Slack channel dedicated to collecting feedback + improving our agents' model configs, skills, and prompts along the axes we care about. so we decided to embed this into our product! every factory @warpdotdev comes built-in with a functional self-improvement loop: > a default set of scorers that judge agent runs on things like code quality and verbosity > over time, scorer results bubble up into a set of concrete improvements to agents' base config waitlist is open:

Warp@warpdotdev

2026.08.18 14:54

Measure factory performance with cost and velocity metrics, accessible from the factory dashboard and via our API and SDK. Plus, configure self-improvement agents to grade performance and automatically improve your factory overtime.