Teaches modern reinforcement learning through hands-on experiments, progressing from foundational concepts to LLM alignment, RLVR techniques, and advanced agentic systems with online training.