As a Research Scientist on the Atlas Perception team, you will play a key role in solving challenging perception and machine learning problems for the most advanced humanoid platform in the world. You will join a world-class research team working on scene understanding, object manipulation, and vision foundation models. As a leader in the field you will shape novel machine learning techniques to expand the ability for Atlas to effectively interact with the world.

How you will make an impact:

• Identify, propose, and lead research projects solving real problems using machine learning
• Creatively combine state-of-the-art open-source models and in-house data to advance the robot capabilities
• Write high-quality (documented, tested, maintainable) Python and C++ code in a large shared codebase
• Regularly present your research within the Atlas team and in wider company meetings
• Collaborate with controls, planning, software and hardware teams to test and integrate your ideas on Atlas

We are looking for:

• Strong ML / Computer Vision / Computer Graphics research background
• Track record of published peer-reviewed research in topics such as:
• Learned 3D reconstruction, depth prediction, feature matching, semantic segmentation, object 6-DoF pose and shape estimation
• Vision-language models
• Multimodal perception with vision/proprioception/touch sensing
• Learned visual representations and foundation models for robot manipulation
• Proficiency in Python programming and popular ML frameworks

Additional desired experience:

• PhD in Computer Science, Machine Learning, Robotics, Computer Graphics or a related field
• Hands-on robotics experience
• Practical knowledge of 3D geometry, projective geometry, camera models
• Competent with C++ programming

LI-JM

Please refer the Job description for details