Product Evals: A Travel Planner Taste Test
A product note on evaluating an AI travel planner: itinerary quality, OOD scenes, long-context consistency, recommendation taste, and user loops.
Agent
Context Engineering: The Shape of Evidence
A note on RAG and context engineering: retrieval quality, evidence shape, memory boundaries, and why context is a product surface.
Agentic RL: The Long Shadow of Feedback
A field note on agentic RL: reward design, behavior shaping, credit assignment, online and offline evaluation, and the feedback loops behind agents.
Evals As Instruments: What Demos Hide
A note on evaluation as an instrument: failure cases, metrics, benchmark design, product loops, and the discipline of measuring agents.
Agent Design: Loops, Tools, and Memory
A field note on designing agents as observable loops, with tools, memory, failure recovery, and product boundaries.