Robin Salimans’ Post

🧠 Agentic AI systems are only as good as the environments they learn in. Join Will Brown from Prime Intellect and me coming Tuesday for a chat about RL Environments! We’ll go under the hood of how we’re building reinforcement learning (RL) environments for agentic AI systems, and how verifier-driven evaluation helps us make those systems more reliable and scalable. Here's what we'll talk about 👇 1. What are RL environments? A look at how we use controlled, instrumented settings to let agents safely learn from feedback. 2. Building RL environments at Zapier How we’re applying the Verifiers library and GEPA (Genetic-Pareto) optimization to evaluate and refine agent behavior on real automation tasks. 3. Lessons learned so far Early insights into what’s worked, what hasn’t, and how we’re evolving our approach to agent training and evaluation. We'll also have some cool surprises for you, so make sure to join! 😏 If you’re interested in RL, verifiable evaluation, or applying agentic methods in production, then this should be a good session. 🎙️ Under the Hood: RL Environments at Zapier 🗓️ Tuesday, December 2, 1 PM ET Link to register in the comments