MTS – Robotics and Physical AI (AI2523) - San Jose, California, United States

Originally posted 14 July 2026 by the employer.

About the role

This role focuses on building vision-based perception, localization, and navigation systems for autonomous mobile robots, drones, and robotic platforms. You will deploy algorithms across traditional geometric vision methods and deep-learning approaches on embedded and edge-compute platforms.

What you'll do

- Architect and develop vision-based perception, localization, mapping, and navigation systems for AMRs, drones, and other robotic platforms. - Develop and integrate visual odometry, visual-inertial odometry, visual SLAM, sensor fusion, and 3D scene-understanding pipelines. - Apply traditional computer vision and deep-learning techniques for feature extraction, object detection, segmentation, depth estimation, tracking, and obstacle understanding. - Optimize and deploy computer vision and AI workloads on embedded and edge-compute platforms. - Collaborate with robotics, hardware, AI, and systems teams to validate solutions in simulation, hardware-in-the-loop environments, and real-world deployments. - Provide technical leadership, mentor engineers, and guide robotics and Physical AI initiatives.

What you'll need

- Master’s or Ph.D. in Robotics, Computer Vision, Computer Engineering, Electrical Engineering, Computer Science, or equivalent practical experience. - 8+ years of experience developing perception, localization, or navigation systems for autonomous robots. - Strong proficiency in C++, Python, and ROS or ROS 2. - Deep experience with visual odometry, visual-inertial odometry, visual SLAM, multi-view geometry, camera calibration, and sensor fusion. - Strong foundation in traditional computer vision, including feature detection and matching, optical flow, epipolar geometry, pose estimation, triangulation, and bundle adjustment. - Experience deploying vision and AI workloads using frameworks such as OpenCV, PyTorch, or TensorFlow on embedded Linux or edge-compute platforms.

Nice to have

- Experience with drone navigation, AMRs, or other GPS-denied autonomous systems. - Experience with stereo vision, monocular depth, dense mapping, neural reconstruction, occupancy estimation, or BEV-based perception. - Familiarity with transformers, vision-language models, diffusion models, world models, or vision-language-action systems. - Experience with simulation and testing platforms such as Gazebo, Isaac Sim, or others. - Demonstrated innovation through patents, publications, or deployed commercial systems.

Skills: Robotics, Physical AI, vision-based perception, localization, navigation systems, autonomous mobile robots, visual SLAM, embedded platforms