awt_akris[asciilifeform]: Not a terrible summary: https://paste.rs/cVk10.md. Seems to be confused about a tangentially related debugging discussion.
signpost[asciilifeform]: might need to chew the logs into embeddings to get good RAG over them. maybe the MCP server already does this?
awt_akris[asciilifeform]: signpost: don't know what you mean be embeddings.
awt_akris[asciilifeform]: @!ask In the context of LLMs and RAG what are embeddings?
raip[asciilifeform]: awt_akris: https://paste.rs/8XG6H.md
signpost[asciilifeform]: typical search poopline would involve processing the messages using an embedding model, obtaining a vector which is a stab at representing the text's position in a vector space learned by the embedding model, storing that embedding alongside the text
signpost[asciilifeform]: then when you want to query, having an LLM elaborate the user's query a bit (to produce more text that has a chance of being similar to embeddings in your db), and then using e.g. cosine distance to rank database messages by similarity, taking the top however many, then feeding those to an LLM to synthesize a response
signpost[asciilifeform] yaps further to kick off another backlog sync
signpost[asciilifeform]: (implemented prod on boot so I don't have to keep doing that)
awt_akris[asciilifeform]: Ok started generating embeddings. Responses seem a bit more thorough.
awt_akris[asciilifeform]: This is a pretty good result for "What programming language was jonsykkel's pest station written in": https://paste.rs/h3MPz.md
signpost[asciilifeform]: but what language is jonsykkel's website written in?
signpost[asciilifeform]: nobody knows
awt_akris[asciilifeform]: old norse
jonsykkel[asciilifeform]: worx gr8 u shud try
jonsykkel[asciilifeform]: old norse is spiritual predecessor to php