Built an autonomous AI RAG Agent with long-term memory using n8n, Supabase, and Postgres

I built an autonomous AI RAG Agent with long term memory using n8n, Supabase, and Postgres. This architecture handles complex, long-context data extraction:

Memory: Integrated Postgres Chat Memory to give the LLM ongoing context across sessions.

Knowledge Base: Connected a Supabase Vector Store utilizing OpenAI Text Embeddings for accurate, real-time semantic search.

Ingestion: Automated a pipeline that dynamically pulls external files from Google Drive and structures them into vector embeddings.

I tested it by asking it "who is Destiny," and it pulled accurate details straight from my own resume, no hallucinating.

I'm sharing this because I'm looking for an AI Automation Internship, and wanted to lead with proof of what I can actually build rather than just a resume.

Founders: if you're spending manual hours on operations, data parsing, or customer workflows, this is the kind of thing I build. I'd love to help automate that for you and save your team real time every week. DM me, I'm ready to start immediately.

#n8n #RAG #AIAutomation