Help , RAG data update

[Pichayut_Prasertwit]

My current RAG flow is:

1. Upload files to Google Drive

2. n8n downloads → chunks/embeds → upserts to Pinecone

3. Chatbot queries Pinecone

But what’s the best way to handle deletes/updates?

• If a file on Drive is deleted, how do I remove its vectors from Pinecone?

• If a file is updated, should I delete old vectors first, then re-embed?

• Do I need to wipe and rebuild the whole index/namespace every time there’s a change, or is there a better pattern?

Any examples or node setups in n8n would help a lot. Thanks!

[moosa]

One solution could be this.

@Pichayut_Prasertwit

If a file is updated, should I delete old vectors first, then re-embed?

or you can try using supabase:

RAG Pinecone (oro any vector store) + Updating Documents

Hi @Michael_Barnett as far as I know this is the common way with upserting vector stores. You basically want to use metadata as much as possible to granulate the chunks, in order to minimize what is being deleted and re-inserted when updating. Our community member @octionic made a pretty neat Youtube video about that if you want to check it out (he uses Supabase though but the concept is the same) [Upsert huge documents in a vector store]