Automated Vector Database Loader from Google Drive

This n8n workflow automates the process of managing and embedding documents stored on Google Drive into a vector database for efficient retrieval and analysis. It is designed for use cases like AI-powered document search, knowledge base updates, or data analysis workflows. The workflow starts with a manual trigger or scheduled interval, searches Google Drive for relevant files, downloads each file, and categorizes it based on its MIME type. PDF files are extracted for text content; text and JSON files are processed accordingly. Extracted text is split into manageable chunks, embedded via OpenAI’s API, and stored in a PostgreSQL vector database, enabling advanced search capabilities. Files are then organized into a specific Google Drive folder, completing the data ingestion cycle.