Document Q&A RAG System: OpenAI, Pinecone & Google Drive

This n8n workflow creates an AI-powered Document Q&A system using Retrieval-Augmented Generation (RAG). It automates ingesting PDF documents from Google Drive, splitting them into chunks, generating vector embeddings with OpenAI's text-embedding-3-small model, and storing them in Pinecone Vector DB for semantic search. A second flow (via webhook) enables natural language querying, retrieving relevant chunks and generating answers with OpenAI GPT.

Key benefits include effortless document management, accurate semantic search over large files, and integration with custom UIs like Lovable for chat interfaces. It saves hours of manual research by providing instant, context-aware responses to queries on contracts, HR policies, manuals, or financial docs.

Ideal use cases: Finance teams querying contracts, legal reviews, HR policy lookups, or any knowledge base needing AI search. Compatible with self-hosted n8n using community nodes; supports batch processing with chunk size 1000/overlap 100, Pinecone insert mode in us-east-1.