AI Evals AI
Hi, I'm Kannan — a quality engineer working on AI evaluations, agents, and MCP testing. This is where I write about making AI systems reliable: how to evaluate them, where they break, and what quality means when the output is non-deterministic.
Latest posts
What is an eval, really?
A short, practical definition of AI evaluations and why they feel different from ordinary software tests.
Hello, world — the pipeline works
A first test post to confirm the domain, repo, and Cloudflare deploy all work end to end.