Max Robotics — Global Service Robot Directory + US Certification

About

Diego Reyes is a leading researcher at the intersection of robotics, natural language processing, and embodied AI. His work fundamentally addresses how robots can bridge the gap between high-level human commands and low-level physical actions. Reyes is best known for his seminal 2022 paper, "Do As I Can, Not As I Say," which introduced the concept of grounding large language models in robotic affordances. This highly influential work, with over 516 citations, demonstrated how to constrain a language model's semantic knowledge with a robot's actual physical capabilities, enabling more practical and safe task execution. More recently, Reyes has pushed the boundaries of legged locomotion with his 2023 work on "Barkour," a benchmark designed to quantify animal-level agility in quadruped robots. This framework provides a standardized way to evaluate complex, dynamic maneuvers like sprinting and leaping, moving the field beyond simple walking gaits. By combining linguistic understanding with robust physical control, Reyes is shaping a future where robots can both comprehend complex instructions and execute them with unprecedented dexterity and speed.