Zeyang Li is a graduate student in mechanical engineering advancing the design of reinforcement learning algorithms that are provably safe, robust, and scalable for real-world control systems. His research integrates control theory and machine learning to ensure that AI-driven agents, such as autonomous vehicles or collaborative robots, can make intelligent decisions without compromising safety. As a second-year MathWorks Fellow, Zeyang will develop frameworks for safe reinforcement learning and data-driven safety filters. He uses computational methods to simulate dynamical systems, implement control algorithms, and solve optimization problems. He has built and validated algorithms that provide strong theoretical guarantees while remaining practical for complex tasks. Zeyang’s work is helping to lay the foundation for intelligent systems that operate reliably in uncertain, safety-critical environments, unlocking new capabilities in autonomous technology and large-scale decision-making.