
Building dexterous robots that learn from human biomechanics
About the role
You will develop machine learning systems that enable robots to acquire dexterous manipulation skills from human demonstrations and multimodal sensor data. You will work across data processing, model training, evaluation, and real-world deployment, using experiments to understand what makes robots learn more effectively. Our research involves vision, motion, biomechanical signals, and robot interaction data, with applications to imitation learning, vision-language-action models, video models, and world models. We are looking for engineers with strong machine learning fundamentals and the ability to turn research ideas into working systems. Prior robotics experience is not required.
What you’ll do
What we’re looking for
Working at Relari
Relari is a small research and engineering startup developing new ways for robots to learn dexterous skills from human biomechanics. Our founders have AI research roots at MIT and NVIDIA, along with autonomous-vehicle and robotics deployment experience at Pony AI and Dexterity. We are backed by top investors including Y Combinator, General Catalyst, and Soma Capital. You will work directly with the founders, own problems end to end, and test your ideas on real robotic systems. This role is full-time and on-site in San Francisco.
Relari is a robotics research company developing new ways for robots to learn dexterous manipulation from humans. We combine video with rich, scalable biomechanical data—including electromyography (EMG), motion, force, and touch—to train robotics foundation models for more capable, human-like manipulation.