Transform Google Drive Docs to Vector Embeddings

This n8n workflow automates the transformation of documents from a Google Drive folder into query-ready vector embeddings. It triggers on new files (PDF, TXT, JSON), downloads them, splits text using LangChain, generates embeddings with OpenAI's text-embedding-3-small model, and upserts them into a Postgres database with PGVector extension. Processed files are then moved to a 'vectorized' folder to prevent reprocessing.

Key benefits include seamless integration of multiple services without custom code, deduplication, and scalability for daily scheduled runs (default 3 AM). It saves hours of manual data ingestion and enables advanced AI features like semantic search across private docs.

Use cases: Power RAG-based AI agents, build knowledge bases for chatbots, enable document classification, or index files for enterprise search. Requires Google Drive OAuth, OpenAI API key, and Postgres PGVector setup.

Highlights: Manual/scheduled triggers, multi-file support, CC BY-SA 4.0 license by AlexK1919.