Build a document QA system with RAG using Milvus, Cohere, and OpenAI for Google Drive

Template Description

This template creates a powerful Retrieval Augmented Generation (RAG) AI agent workflow in [n8n](https://n8n.partnerlinks.io/5xf5bs8y3ruv). It monitors a specified Google Drive folder for new PDF files, extracts their content, generates vector embeddings using Cohere, and stores these embeddings in a Milvus vector database. Subsequently, it enables a RAG agent that can retrieve relevant information from the Milvus database based on user queries and generate responses using OpenAI, enhanced by the retrieved context.

Functionality

The workflow automates the process of ingesting documents into a vector database for use with a RAG system.

1. Watch New Files: Triggers when a new file (specifically targeting PDFs) is added to a designated Google Drive folder.

2. Download New: Downloads the newly added file from Google Drive.

3. Extract from File: Extracts text content from the downloaded PDF file.

4. Default Data Loader / Set Chunks: Processes the extracted text, splitting it into manageable chunks for embedding.