The rapid growth of Generative AI assets introduces new risks and complexity for organizations. This session explores AI observability and model evaluation techniques for Generative AI Large Language Models (LLMs) in production.
We will cover strategies to ensure performance, accuracy, and reliability, through evaluation metrics, guardrails, and user feedback. Participants will learn to implement observability across environments, drawing insights from real-world use cases.
Speakers
Atalia Horenshtien
Scott Munson