Concerned about the quality and reliability of your generative AI models? Join us in this session if you understand the critical role of evaluation in development. Learn from Google and industry experts as we focus on how the Vertex AI evaluation service helps you validate every stage of the life cycle for quality and trust.

Generative AI is moving to production, powering complex workflows. Evaluation is now crucial. This deep dive explores core evaluation principles and methods, with a focus on multi-modal assessment. We'll dissect practical challenges, best practices, and hear from leading researchers on cutting-edge techniques for evaluating multi-modal use cases. Learn how to measure generative AI success beyond simple outputs.

Evaluating the quality of a model’s responses relative to other models is a top challenge in deployment. Vertex AI makes this easier with new evaluation capabilities that help you assign a score to the quality of a model response. In this session, you will learn how to use evaluation tooling, best practices to set up your task-based evaluation framework, and how to use explanations to build trust in your evaluation.

Click the blue “Learn more” button above to tap into special offers designed to help you implement what you are learning at Google Cloud Next 25.