Blog

Prompt engineering without measurement is guessing with extra steps. How to build an eval set from real traffic, pick a grader that isn't lying to you, and avoid tuning your way to a number that means nothing.

Everything an agent reads is input, and a model has no structural way to separate instructions from data. Why mitigations in the prompt are not a boundary, and where the real one goes.

A list of numbers where distance means similarity. What cosine similarity measures, why the same word can have two vectors, and the queries embeddings are quietly bad at.

Most RAG failures are retrieval failures, not generation failures. How chunking and embeddings decide what the model ever sees, the similarity-is-not-relevance trap, and what actually fixes it.

The window is a budget you refill from scratch every turn. Where the tokens actually go in an agent, why prompt caching is a prefix match, and the difference between clearing context and compacting it.

A tool is a name, a description, and a JSON Schema — and the description is prompt text that gets read every turn. Schema design, validation, error messages that repair, and why fewer tools work better.

An agent is a while loop around a stateless API. What stop_reason really controls, why the transcript is resent on every turn, and the failure modes that show up once tools are in the mix.