Why RL Environments are all you need for building agents
Watch the recording
Watch on YouTubeRL Environments are used by frontier labs to train their latest models and agents. In this talk, I will explain why RL environments are fundamental to all stages of building agents: from evaluations and prompt optimization to post-training agents. I will show concrete industry examples of how practitioners use RL environments to evaluate their agents, and to understand how their agents fail. Lastly, I will also talk about how one can set up data flywheels which helps you figure out how to improve your agents on your specific data and use-cases.
About the speaker
Mahesh Sathiamoorthy:
Mahesh Sathiamoorthy is the co-founder and CEO of Bespoke Labs, a data research lab that has raised $40M in Seed and Series A. Previously he was at Google DeepMind where he published seminal work in the intersection of LLMs and recommender systems, which are now being adopted widely in the recommender systems industry across the world.
More Sessions
Join the builders, testers, and innovators shaping the next generation of web experiences.
Testμ Conf 2026 is where they meet.
