Max Robotics — Global Service Robot Directory + US Certification

About

Alex Paino is a leading researcher in robotics and reinforcement learning, best known for pioneering work that bridges the simulation-to-reality gap. His most celebrated contribution is the 2019 paper "Solving Rubik's Cube with a Robot Hand" (632 citations), which demonstrated for the first time that a dexterous manipulation policy trained entirely in simulation could solve a complex, real-world puzzle. This breakthrough was enabled by his invention of **automatic domain randomization (ADR)** , a technique that systematically varies physics parameters in simulation to produce policies robust to real-world dynamics. Paino also advanced the field of goal-conditioned reinforcement learning with his work on **asymmetric self-play** (2021, 21 citations), where two agents—Alice and Bob—automatically discover and propose increasingly challenging manipulation goals, allowing a single policy to generalize to unseen objects and tasks. His research has profoundly influenced how roboticists approach training for dexterous manipulation, reducing reliance on expensive real-world data collection. Paino’s work at the intersection of simulation, self-supervised learning, and robotics continues to inspire new methods for autonomous skill acquisition.