Max Robotics — Global Service Robot Directory + US Certification

About

Sally Jesmonth is a leading researcher at the intersection of robotics, natural language processing, and large-scale machine learning. Her work focuses on grounding high-level language instructions in real-world robotic actions, bridging the gap between semantic knowledge and physical affordances. Her landmark paper, “Do As I Can, Not As I Say” (2022, 516 citations), introduced a groundbreaking framework that leverages large language models to guide robots in executing temporally extended commands, overcoming their lack of real-world grounding. This work has become a cornerstone for language-conditioned robotics. She is also a key contributor to the RT-1: Robotics Transformer series (2022–2023, over 550 combined citations), which demonstrated how knowledge from diverse, task-agnostic datasets can be transferred to enable scalable, real-world robot control with minimal task-specific data. Her research has profoundly influenced how robots learn from both language and experience, achieving high performance in zero-shot and few-shot settings. Jesmonth’s contributions are shaping the future of embodied AI, making robots more capable of understanding and acting upon human instructions in dynamic environments.