Learn
Learn how agents learn from consequences.
Plain-language guides to RL environments, verifiable rewards and reward hacking, written by the team that builds them.
01FoundationsWhat is an RL environment? A guide for LLM agentsAn RL environment is a simulated world where an AI agent takes actions, the world changes in response, and a reward signal says how well the agent did.9 min read 02RewardsVerifiable rewards (RLVR), explainedReinforcement learning with verifiable rewards (RLVR) trains a model with rewards computed by a program that checks the result, such as an answer key or a test suite, instead of a learned reward model.8 min read 03RewardsReward hacking: how AI agents game their gradersReward hacking happens when an AI system scores well by exploiting a flaw in how it is graded, rather than by doing the intended task.9 min read 04BuyingChoosing an RL environment vendor: a buyer’s checklistWhen you license an RL environment you are buying a reward signal as much as a simulator.8 min read