Let's Talk About LLM Evaluations

In this 31-minute conference talk presented at All Things Open AI 2025, John Willis, AI CIO, explores the critical topic of Large Language Model (LLM) evaluations. Dive into methodologies, challenges, and best practices for assessing LLM performance and capabilities. Learn how organizations can effectively measure and compare different language models to make informed implementation decisions. The presentation offers valuable insights for AI practitioners, developers, and decision-makers looking to understand how to properly evaluate LLMs in various contexts and applications.