JJacksoninrecalso.hashnode.dev·Aug 11 · 5 min readWhat actually loads into a Claude Code session, and whenMost advice about CLAUDE.md stops at "keep it short." That is true and not very useful, because it does not tell you where the rest of the material should go. The useful version is a load model: which12S
JJacksoninrecalso.hashnode.dev·Jul 30 · 7 min readYour hybrid retriever is fusing scores that were never on the same scaleMost hybrid search tutorials end at the same place: run BM25, run a vector search, add the two scores with a weight, sort. It works on the demo corpus. Then the corpus grows and the ranking quietly ch00
JJacksoninrecalso.hashnode.dev·Jul 12 · 8 min readHybrid search on-device: why we index every filename with SQLite FTS5 and embed only what mattersThe bug report that killed our "just use embeddings" plan A user knows the file exists. They typed its name. They can picture the icon. And the semantic search we shipped confidently returned four the00
JJacksoninrecalso.hashnode.dev·Jul 6 · 9 min readWhat poisoning a RAG store taught us about agent memoryMost of the "AI memory" conversation is about getting things in: chunking, embeddings, which vector database, how to compress a long chat into something you can recall later. Almost none of it is abou00
