Max Robotics — Global Service Robot Directory + US Certification

About

Keerthana Gopalakrishnan is a robotics and machine learning researcher whose work sits at the intersection of large language models, vision-language systems, and real-world robotic control. She has made significant contributions to the emerging field of foundation models for robotics, helping pioneer approaches that enable robots to understand and act upon complex natural language instructions grounded in physical environments. Her most influential work includes the landmark "Do As I Can, Not As I Say" paper (516 citations), which demonstrated how language models could be paired with robotic affordances to execute high-level tasks, and the RT-1 and RT-2 Robotics Transformer series (512 and 267 citations respectively), which showed that large-scale transformer architectures trained on diverse data could achieve remarkable generalization in real-world robotic manipulation. Her development of NLMap further advanced open-vocabulary scene representations for robotic planning, while RoboVQA tackled long-horizon multimodal reasoning at scale. Collectively, her papers have accumulated over 1,500 citations, reflecting her outsized impact on how robots learn from internet-scale knowledge. Through projects like AutoRT and Q-Transformer, Gopalakrishnan continues pushing the boundaries of scalable, deployable robot learning — making her a central voice in the next generation of embodied AI research.