Published thesis titled 'Learning to Manipulate Using Diverse Datasets' upon graduation from CMU; RoboNet project featured in MIT Tech Review; Received NDSEG fellowship in 2021; Contributed to the Gemini Robotics project aimed at developing AI models for physical-world applications.
Research Experience
Research Scientist at Meta Reality Labs; Worked for a year at Google Deepmind; Interned at Nvidia's robotics lab in Seattle under Ankur Handa and Dieter Fox; Organized the Cloud Robotics and Benchmarking Workshop at RSS 2022.
Education
PhD in Robotics from Carnegie Mellon's School of Computer Science, Advisor: Professor Abhinav Gupta, Specific details on time and specialization not provided.
Background
Research Interests: Intersection of computer vision, machine learning, and robotics. Professional Focus: Internet-scale representation learning, robot learning, reinforcement/imitation learning. Brief Introduction: Aspires to build actionable foundation models that can help robots (and other autonomous agents) generalize in the real world.
Miscellany
Contact: sdasari at alumni dot cmu dot edu; Social Media: GitHub, Google Scholar, Twitter