This is the first post on the Epsilab blog. We'll use this space to share what we're learning while building high-fidelity RL environments for real-world agents.

What to expect

Deep dives on how we build environments and verifiers.
Notes on training and evaluating long-horizon agents.
Product updates as new environments come online.

Have a topic you want us to cover? Reach us at founders@epsilab.com.