@Yeye

Recent activity

I stopped running benchmarks on my agents. Here's what replaced them in a 5-agent shop.

How do you prevent AI coding assistants from forgetting your architecture?

One tester reported a bounded AI-agent run; the useful part was the amendment, not the success claim

One tester reported a bounded AI-agent run; the useful part was the amendment, not the success claim

One tester reported a bounded AI-agent run; the useful part was the amendment, not the success claim

See all activity