Jeremy Flint

I'm a sophomore at MIT majoring in AI & Decision Making. I'm interested in benchmarking and evaluation for AI safety — building better ways to measure what AI systems can and can't do reliably.

This past fall, I was on leave building agents at Shortcut by Fundamental Research.

You can find some of the things I've built here.

Recently, I created an AI-powered skincare app and joined Stitch to design a graph-based LLM interface.

I also enjoy swimming and skateboarding in my spare time.