Agentic RAG, Open LLMs, FREE Embeddings | n8n Tutorial
Build an agentic RAG workflow in n8n using open models, Hugging Face, embeddings, vector search, and MCP tools. In this walkthrough I show how n8n works, how to run it locally or on a VPS, how RAG has changed from old-school retrieval to agent-driven vector search, and how to build ingestion and retrieval workflows inside n8n. --- Links - n8n: https://link.alejandro-ao.com/n8n --- Topics Covered - n8n workflow automation for AI agents - Agentic RAG with vector search tools - Open models with Hugging Face and MCP --- Timestamps 00:00 - Introduction 01:33 - What is n8n 04:18 - Installing n8n 07:35 - RAG ingestion concepts 13:36 - Agentic retrieval 21:23 - Ingestion flow in n8n 32:48 - n8n agentic retrieval flow 47:43 - Observability 49:22 - Use cases and outro ## Connect with me X: https://x.com/_alejandroao LinkedIn: https://www.linkedin.com/in/alejandro-ao/