Kolo is for engineers excited about the potential of AI, but disappointed about coding agents’ actual performance. It’s for folks who want their computer to help wrangle complexity, not add its own.

Yes! While our agent features are waitlist-only, we’re currently helping thousands of humans visualize their Python code, debug it, and generate instant tests. Check out our docs.

We’re working on it! In the meantime, research has shown that runtime traces consistently enhance agent performance by up to 9.8% across the HumanEval, MBPP, and TransCoder benchmarks.

Ben is writing this (hello!), and Wil’s sitting beside me. He also says hi. Previously, Wil built SimplePoll, and worked at Github. Ben was an early Stripe employee, and ran operations for a Sequoia-backed startup.

Not currently, but feel free to drop Ben a note if you’d like to chat more. The best way to follow our progress is to join the waitlist.

Not right now. But we’d love to chat in the future! Join the waitlist to keep in touch and hear about possible future opportunities.

Love it. You can drop us a note or pop into our Discord.