// writing

Notes from the middle of the stack.

Short posts on RAG done right, silent-refresh auth, Firestore realtime, and the parts of production software nobody demos.

RAG that actually works in production

Vector search is the easy part. The hard part is chunking, freshness, and knowing when the model should refuse to answer.

The 401 handler you copy-paste from Stack Overflow will fire N parallel refresh calls the moment your access token expires. Here's the fix.

onSnapshot is easy to add and easy to forget. In a SPA, forgetting means memory leaks and phantom writes.