Chris Mair


Talk: Self-hosted RAG

Cat or CPU?

I am currently working on a proof-of-concept implementation of a Retrieval Augmented Generation (RAG) pipeline to build an assistant for the NOI Techpark’s Open Data Hub1 team.

I gave a talk2 about the technology stack I used to implement RAG using only self-hosted and free software and models.

The video of the talk (~1.5 hours) is available on NOI Techpark’s YouTube channel:

The collection of documents and source code shown during the talk is available on my Codeberg page: