Agent-Driven Development in Cursor: Testing, Benchmarking, and Optimizing Functions

I’ve been dreaming of “Agent-Driven” development for a long time so I thought I’d dive in and see how close we are. The goal was to get a Cursor agent to create all of the unit tests and benchmarks then automatically test/benchmark a better version of current function. Hit a few bumps here and there but otherwise walked away seriously impressed.

3 Likes

Great video @johnlindquist thanks for sharing it. I’m curious about what Voice to Text app you’re using? Looks neat