Varoon Kodithala(@varoonkodithala):this is something i’m building! ingesting user feedback and LLM-as-a-judge evals directly into improvements to agent prompts/skills is absolutely meta. has made huge improvements to the efficiency of our internal factory.

Varoon Kodithala

@varoonkodithala

i run marathons and write code @warpdotdev | previously @contrary, @georgiatech

Joined August 2019

1.2K Following 1.3K Followers

Varoon Kodithala@varoonkodithala

2026.08.25 22:23

this is something i’m building! ingesting user feedback and LLM-as-a-judge evals directly into improvements to agent prompts/skills is absolutely meta. has made huge improvements to the efficiency of our internal factory.

Zach Lloyd@zachlloydtweets

2026.08.25 19:54

This is an example of a productized self-improvement loop on code review in Warp Factories. You can see here that our observer agents are scanning PR comments for patterns and suggesting changes to our factory PR skills that we then review, merge, and improve.