RAG Knowledge Chatbot: OpenAI, Google Drive & Supabase

This n8n workflow creates a Retrieval-Augmented Generation (RAG) chatbot that ingests PDFs from Google Drive, chunks and embeds them using OpenAI, and stores vectors in Supabase for semantic search. When users ask questions, it retrieves relevant document chunks, augments OpenAI prompts, and generates precise answers with optional PostgreSQL chat memory for conversation continuity.

Key benefits include eliminating manual document searches, providing instant answers from proprietary knowledge bases, and scaling effortlessly with cloud services. It's ideal for reducing response times in customer support, sales teams accessing proposal templates, or internal teams querying policies and guides.

Use cases: Sales reps generating tailored business proposals from product docs; support agents resolving queries from FAQs/manuals; researchers or educators building interactive Q&A from papers. Setup requires minimal configuration after importing, making it accessible for non-technical users while handling complex AI orchestration.

The workflow separates ingestion (scheduled or manual) from real-time chat (webhook-triggered), ensuring fresh knowledge without downtime. Enhance productivity by 5-10x in knowledge-intensive tasks.