In this session, Landry Noubissie will provide an in-depth look at the fundamentals of language models like ChatGPT, offering a foundation on how these models work.

Landry will explore the differences between CPU and GPU processing (AMD vs. Nvidia) and guide us through using the Transformer module in Python to implement various AI models, including Llama and IBM’s Granite 3.0 open-source models. He’ll demonstrate how to run AI models locally, ensuring data privacy and security without relying on external APIs.

The session will cover prompt engineering basics and introduce AI agents, focusing on their implementation in Python. Landry’s talk is particularly suited for businesses with limited budgets seeking to maintain data privacy by setting up and managing AI tools locally, using freely available resources.

Speaker: Landry Noubissie