After chucking the first version of the eval harness for Zoltar, I started over with a minimal version that I could easily understand. I've learned that an LLM judge can tell me if a pass rate improved, but it can't yet tell me if the pass rate is good enough.

I didn't realize I was vibe-coding my project until I tried to write an article about its testing infrastructure and couldn't. The solution turns out to be writing documentation by and for humans.

Claude Code logs every session locally in an undocumented JSONL format. I wrote cctx to turn those logs into markdown transcripts you can commit and review alongside the code, so the reasoning behind a decision is preserved and not just its outcome.

You can't stop an AI feature from hallucinating, but you can keep nonconforming data out of production documents by storing state in an authoritative database, structuring updates with tool-calls, and validating data before it reaches the database.

Building a reliable system around an LLM means dealing with its weird failure modes. The conventional wisdom on retries doesn't fit — a second failure isn't a model problem. It's a signal that the validator, prompt, or schema needs work.

When searching a body of text, the default reflex is to reach for a hybrid search that combines full-text search (FTS) with semantic searches. On a personal corpus, hybrid search consistently underperforms one of its components; FTS is usually but not always the right default.