for those of you who are autoresearch pilled , or have been meaning to get into autoresearch but dont know how - I shipped evo today - a opensource Claude Code plugin that optimizes code through experiments
you hand it a codebase. it finds a benchmark, runs the baseline, then fires off parallel agents to try to beat it. kept if better, discarded if worse.
inspired by @karpathy's autoresearch, but with structure on top:
- tree search over greedy hill-climb — multiple forks from any committed node
- N parallel agents in git worktrees
- shared failure traces so agents don't repeat each other's mistakes
- regression gates
2996 customers later, it’s time to release ScreensDesign V2 !
3 months of work.
A more complete library.
A truly agentic /create.
→ Research what works in real iOS apps
→ generate onboarding, paywalls, and full app flows
→ hand it to AI coding agents
→ Make the printer go brrrr
First 200 retweets+replies/DMs get free credits dm'ed ;)
https://t.co/iVMh45gGSu
@MilksandMatcha I am raising two toddlers, working a full time job overnight security at a hospital, running and operating a gym/personal training business, and finishing my first year of Physical Therapy school. I’m using codex to run my business, build a personal tutoring program, and schedule
Every day we wait, they suffer.
Right now, there are kids praying for a rescue—for someone to step up, step in, and do something.
Are we actually going to protect these kids from further exploitation, or are we just going to keep talking about it?