Running AI systems often feels like it demands endless resources—powerful servers, constant energy, and hefty budgets. But it doesn’t have to be that way. In this session, we’ll explore how to use local large language models (LLMs) on small-footprint hardware to create systems that are efficient, effective, and easier on the planet.

We’ll cover the technical basics of deploying local LLMs, look at how they stack up against cloud-based solutions, and walk through examples of how others are making this shift. You’ll leave with practical ideas you can apply to your own IT projects, helping you reduce energy use and cut costs without sacrificing performance.

This talk isn’t about fancy ideas—it’s about simple, solid steps toward a more sustainable approach to technology. Let’s rethink what “smart systems” can really mean.