Claude is building agent memory functionality into its products, and Adam's team uses separate running memory markdown files that Claude updates, which is more effective and lightweight than RAG because it avoids the overhead of maintaining vector databases.

causalpending

Speaker

Adam

Evidence Quote

this is much more effective than rag because rag runs based on semantic search, but it's it's also overkill really to have to set up a vector database and maintain that vector database

Source

Karpathy's LLM Wiki Goes Further Than Everyone RealisedAdam Goodyer
Created: 8/13/2026, 9:47:09 AM

My Notes

Loading notes...