AI-Powered Two-Stage Document Retrieval Chatbot with OpenAI & Supabase

Overview

This workflow implements a two-stage document retrieval chatbot that leverages OpenAI's language models and Supabase's vector search capabilities. It enables users to interact with a chat interface, retrieve relevant documents, and receive AI-generated responses based on document content.

Key Features

• Chat Trigger: Listens for incoming chat messages to initiate the workflow.
• Document Retrieval: Uses Supabase vector search to find and filter documents based on semantic similarity to the user's query.
• AI Integration: Employs OpenAI for both embedding creation and generating conversational responses.
• Automated Embedding Updates: Keeps document embeddings up-to-date for accurate search results.
• Aggregation & Filtering: Aggregates and filters files to ensure only the most relevant documents are used in responses.

Benefits

• Enhanced Productivity: Automates the process of searching and summarizing documents, saving significant time.
• Accurate Responses: Delivers contextually relevant answers by combining vector search with AI-powered language understanding.
• Seamless Integration: Connects Supabase and OpenAI within n8n for a unified, low-code automation solution.

Use Cases

• Customer support chatbots that reference internal documentation.
• Knowledge management systems for quick information retrieval.
• Automated assistants for HR, sales, or project management teams.