Hands-on AI testing. Evidence for the SI era.

Models, agents, tools and real-world workflows. We run the experiment, show the evidence, keep the failures, and publish what actually works.

Lab reports