#ai testing

RAG Triangulator

You're not chatting with an AI - you're investigating why one gave a confident, wrong answer. Inspect a simulated retrieval pipeline, change Top-K and context budget, and prove where the failure actually happened.

What I Learned Building a Testing Framework for a Local AI Engine

A real story about building a testing framework for a local AI inference engine, and what it taught about verifying assumptions before trusting your own tests.

I Asked a Local AI Agent to Add 7 and 5

A hands on look at whether two small local models could actually call a tool in Hermes Agent, and why the database told a different story than the terminal screen.