RUG::B

Everything today seems to be about LLMs, and this talk is no different! We'll dive into the basics of Retrieval-Augmented Generation (RAG), a clever way to make large language models work with your own data. RAG enables AI systems to pull relevant information from your unique content, allowing for smarter, more tailored responses. Join us to explore how this concept is shaping new possibilities for data-driven, context-aware applications.

Premature optimization is the root of all evil, so the saying goes, but have you ever endured the hell that is waiting for a slow program to complete?

Let's learn how we can break with the preconceived notions of optimization, learn what modern hardware can do for us and how we can write ruby programs that work with the hardware and not against it.

====

Buyer beware:

This is an enhanced version of a talk I gave last September: https://www.rug-b.de/topics/performance-it-can-t-be-an-afterthought-1022