Building an AI Assistant with my own Knowledge Base

Questions
1

This is one of the most desired AI projects and I know there are specific tools for this. But since n8n can handle so much, why not use it for one more job :muscle:

I’m having a hard time trying to understand how everything works, but the following fluxograms helped me a lot:
image

image

Now my main questions are:

  1. Is it a good idea to set up a local vector store using n8n nodes, or should I use Pinecone instead?

  2. Should I use RAG (Retrieval-Augmented Generation) method to develop an assistant that can generate written content for social media? Or is there something better?

I’ve looked through various workflow templates but I would greatly appreciate any additional insights or tips before I start building this. Thank you!

Information on your n8n setup

  • n8n version: 1.27.2
  • Database (default: SQLite): SQLite
  • n8n EXECUTIONS_PROCESS setting (default: own, main): own, main
  • Running n8n via (Docker, npm, n8n cloud, desktop app): docker
  • Operating system: ubuntu 22.04
3

I’m currently also looking into this! You can already find tons of information on youtube: https://www.youtube.com/results?search_query=N8N+langchain+

I have seen there are already some basic templates available on N8N if you search for langchain nodes, these are basically the AI agents you are looking for and you want to connect to generative GPT libraries to complete specific tasks, connect to your knowledge base and sql (or other ) databases…

4

Here’s a great tutorial by @Derek_Cheung on this topic too:

2 Likes
5

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.