Robot Learning Intern

The role

Robot Learning Interns will work on learning-based methods to make robots more dexterous. You will develop and test new algorithms for manipulation, navigation, and control using simulation and real Dexmate robots.

What you will do

• Develop new algorithms and methods for training AI models that enhance robot dexterity.
• Conduct cutting-edge research across multiple disciplines.
• Design and implement state-of-the-art learning-based manipulation, navigation, and control algorithms on real robots.
• Work with other teams to develop a diverse set of robust manipulation skills for robots.

Required qualifications

• Currently enrolled in a PhD program or have a master degree in Computer Science, Robotics, Electrical Engineering, Mechanical Engineering, or a related technical field.
• Passionate about working with robots.
• Research experience in embodied AI, robotics, computer vision, machine learning, human-AI interaction, or computer science.
• Experience with deep learning frameworks such as PyTorch.
• Solid understanding of SOTA robot learning techniques (reinforcement learning, imitation learning, etc.).
• Experienced with robot simulators such as Isaac Gym/Isaac Sim/SAPIEN/MuJoCo/Drake, etc.
• Experience building systems based on machine learning and/or deep learning methods.

Nice to have

• A track record of research, with work published in top conferences and journals such as Science Robotics, IJRR, RSS, CoRL, ICRA, NeurIPS, ICML, ICLR, CVPR, etc.

About the company

Dexmate is building the foundation for physical AI by combining a new generation of robots with a universal Physical AI OS. Their mission is to democratize robotics by delivering a plug-and-play platform for developers, researchers, and enterprises.