Self-Hosted AI Stack

Custom AI deployment infrastructure: OpenWebUI as auth/frontend, Qdrant for vector RAG, Neo4j for graph RAG — combined into a unified retrieval pipeline running on Hetzner. No cloud LLM dependency.

Full write-up coming soon.

Other projects

01 Coming Soon

Coffee & Code Blog

A space for writing about the intersection of software craft and coffee culture. Long-form, no filler.

02 In Progress

AI Coffee Pairing

Using LLMs to suggest coffee origins and brewing methods based on your flavor profile and preferences.