Fail-closed groundedness, deterministic corroborators, and a self-test — because an evaluator should...
A RAG evaluator that admits what it can't judge
Related
Why I Built xAgent
I started building xAgent in April 2025. The original idea was straightforward: build a...
I Edited the Task Mid-Flight. The Agent Stopped Instead of Guessing.
Permission levels decide what an agent may do. They do not decide what happens when the task itself changes mid-flight. Here is the stop con
Building OopsCalorie: When Your AI Thinks Dinuguan Is Champorado 😂
I’ve been building a side project called OopsCalorie, an AI-powered calorie and meal tracking...