AI-Powered Retrieval System for Document Management and Chatbot

This advanced n8n workflow integrates document processing, vector storage, and AI-powered chat capabilities to create a sophisticated retrieval-augmented generation (RAG) chatbot. It automates the extraction, indexing, and retrieval of documents from Google Drive, enhancing user interactions with contextually relevant responses. The workflow begins by fetching files from a specified Google Drive folder, then processes and splits these documents into manageable chunks. Metadata is extracted for improved search and indexing, and vectors representing document content are stored in Qdrant, a high-performance vector database.

The workflow also includes mechanisms for maintaining and updating the vector store, including secure deletion with human oversight—triggered via Telegram notifications. For interaction, it leverages Google Gemini (PaLM) to power an AI chat model capable of understanding and responding to user queries with reference to stored documents. Users can initiate conversations through webhooks or chat triggers, with chat history stored and managed in Google Docs. This setup is ideal for organizations that need an intelligent, scalable system to manage large document repositories and facilitate precise, context-aware chatbot interactions.