#knowledge-management (4)

Why an LLM-maintained wiki accumulates understanding where retrieval-augmented generation re-derives it on every query.

Karpathy's pattern for an LLM-maintained wiki: instead of re-deriving answers from raw documents on every query (RAG), the model integrates each source into a persistent, compounding knowledge base.

Karpathy's pattern for an LLM-maintained wiki: instead of re-deriving answers from raw documents on every query (RAG), the model integrates each source into a persistent, compounding knowledge base.

A persistent, interlinked knowledge base that a language model builds and maintains — integrating each source once, rather than re-deriving answers from raw documents on every query.