Pacific Design/ artificial intelligence

section 05 · status: live · 4 entries · updated 2026-08-27

RAG & Embeddings

Models know what they read during training and nothing since. Retrieval fixes that: meaning as vectors, similarity search at scale, the pipeline that feeds documents into context, and the honest comparison with fine-tuning.

live · a query landing in embedding space, pulling neighbors into context

  1. Embeddingsmeaning as a point in space
  2. Vector searchfinding neighbors without checking everyone
  3. The RAG pipelinelook it up before answering
  4. RAG vs fine-tuningknowledge in context, behavior in weights