Paul Michaels - Lemmy, Llamas, and Large Language Models Locally

LeedsSharp are proud to present Paul Michaels in "Lemmy, Llamas, and Large Language Models Locally"

Access Details Please go to the Enablis offices, they are on the first floor of the building, both front door and office doors should be open. If there is a problem then press the buzzer. We will meet at 6pm for pizza and drinks, the talks will start at 6.45. Some of us go to the either the brewery tap or editors draft pub afterwards at 8pm (vote to be taken).

Abstract:

In this session, we’ll step back from hype and explore what’s really happening when you run a language model, demystify some of the terms, and play fun games. We’ll look at how modern LLM runtimes expose simple HTTP endpoints, and how you can easily programmatically (in C#) interact with a local model (free).

We'll talk about what an MCP is, and create a server. We'll cover why this is useful (especially locally) and how a language model act as a decision engine that selects structured tools while your .NET application remains firmly in control.

Bio:

Paul Michaels is a Microsoft MVP and a Head of Technology, working across Azure and AWS. He's been programming professionally since 1997, and likes finding neat solutions to problems.