From Big Model to Small Model: Distill Your Own Agent LLM

🗓️ Agenda
18:00 – Welcome & Networking
18:30 – Introduction
18:35 – Presentation/Demo
19:45 – Survey, Quizz & Feedback
20:00 – Closing

🧠 About the Session
Every company using AI agents today faces the same problem: they're paying per-token for a large cloud model, sending sensitive data to third-party APIs, and locked into someone else's infrastructure. There's a better way!

In this session, we will walk through the complete pipeline for building AI agents that run on your own infrastructure, on your own model. You'll see how to capture conversations from a large model like Claude or GPT-5, use that data to fine-tune a smaller open-weight model through distillation, quantize it to run on modest hardware, and deploy a fully functional AI agent on top of it all with no API keys, no per-token costs, no data leaving your environment.