.png&w=1536&q=75)
Free Lesson
How OpenAI Customers Use Evals To Build Better AI Products
Part of The AI Evaluation Handbook
30 min
Jun 9, 2025 3:30 PM
Virtual (Zoom)
In this video
What you'll learn
Case studies from OpenAI customers
Explore real-world case studies showing how evals drive iterative improvement, prevent regressions, and reduce risk.
Get a sneak peek of OpenAI's integrated eval products
OpenAI offers eval products. Learn what they are and how to use them.
The hidden risks of skipping evals
Learn what happens when you don’t do evals: amplified bugs and bias, credibility loss, and internal confusion.
Design patterns for eval-driven development
Get practical techniques for incorporating evals into your workflow—from prompt evaluation to regression testing.
Why this topic matters
Evaluation-driven development is the key to accelerating AI product development. This session will cover data-driven methods in driving model performance and user outcomes. He will also share some features within OpenAI’s platform for integrating evals into AI product workflows.
You'll learn from
Jim Blomo
Engineering Leader at OpenAI, working on Evals
Hamel Husain
ML Engineer with 20 years of experience