Max Robotics — Global Service Robot Directory + US Certification

About

Nimrod Gileadi is a leading researcher at the intersection of large language models, reinforcement learning, and robotic manipulation. His work focuses on enabling robots to learn complex, dexterous behaviors—from piano playing to multi-fingered grasping—by bridging the gap between high-level language understanding and low-level motor control. Gileadi’s major contributions include pioneering the use of LLMs to translate natural language into reward functions for robotic skill synthesis, as detailed in his highly cited work *Language to Rewards for Robotic Skill Synthesis* (2023, 38 citations). He also co-created the MuJoCo MPC framework (*Predictive Sampling*, 2022, 12 citations), an open-source platform for real-time predictive control that has become a standard tool in the field. His work on *RoboPianist* (2023, 5 citations) demonstrated the power of deep RL for dexterous piano playing, while *DemoStart* (2025, 2 citations) introduced a demonstration-led auto-curriculum method for sim-to-real transfer with multi-fingered hands. Gileadi’s research is characterized by its practical impact, open-source contributions, and elegant integration of language, learning, and control—making him a key figure in the next generation of autonomous robotic systems.