Robotics Engineering Intern — AI Manipulation & Perception

About the Role

Work alongside our engineers — directly with the CTO and founding engineers — to build and deploy AI-first mixed-case palletizing on KUKA robots: robots that pick a stream of never-before-seen packages and stack them into pallets and roll-cages, with no per-box programming.

You'll take real ownership of a piece of the stack and see it run on a live cell.

This internship may include travel to some of the world's most exciting factories — Midea, Mercedes-Benz, SF Express, and others — mostly in southern China.

What You'll Do

• Help build and test 3D perception (segmentation, 6-DoF pose) on real depth-camera data.
• Build data pipelines — recording, labeling, and dataset conversion for model training.
• Run bench and simulation tests — camera calibration, gripper trials, and the robot simulator.
• Support on-site work — fake-parcel testing, data collection, and debugging.

What We're Looking For

• Pursuing a BS / MS / PhD in Robotics, Computer Science, EE, or a related field.
• Solid Python, with some computer vision, machine learning (PyTorch), or robotics from coursework or projects.
• Hands-on and self-driven; excited to work with real robots and real deployments.
• Proficient English is required (Chinese is a plus).
• Comfortable using agentic coding tools like Claude Code.
• Bonus (any is a strong plus — this role sits next to a demanding full-time system): 3D perception — class-agnostic segmentation, 6-DoF pose estimation, point clouds, structured-light / depth 3D cameras.
• Bonus: grasping — suction / vacuum end-effectors and grasp selection across varied packaging; placement — online 3D bin-packing / palletizing, stack stability, and load density.
• Bonus: motion & control — motion planning, kinematics, trajectory interpolation, and real-time control; you've worked with a robotic arm (KUKA or RSI a plus).
• Bonus: cloud teleoperation — driving a robot over the network (kinematics, interpolation, latency handling).
• Bonus: NVIDIA Isaac Sim, ROS 2, cuRobo / OMPL, LeRobot, imitation learning / VLAs, or NVIDIA Jetson.
• Bonus: you've built multi-agent systems, or have deep experience with agentic coding tools like Claude Code.

Why Join Us

Work directly with the CTO and founding engineers on a real product going live with world-class industrial partners — KUKA, Midea, Mercedes Benz, and SF Express. You'll get direct advising from leading faculty at UC Berkeley BAIR and from researchers at Physical Intelligence and other company partners. The team feels like a family: Friday-night events and occasional retreats — racing, beach days, and more (last summer we took the whole team to Hawaii!). Some interns are encouraged to pursue open-ended exploration, to better align their skills with the direction of our research. On-site preferred (Shenzhen / Suzhou or Bay Area); full-time, paid internship, with potential to convert to a full-time role.