9 hours ago
Responsibilities
- Build and improve computer vision pipelines for detection, classification, tracking, and temporal understanding.
- Test first-person perception systems under camera motion, changing viewpoints, occlusion, motion blur, and variable lighting.
- Use synthetic and real-world data to identify failure cases and improve model performance.
- Train, evaluate, and debug computer vision models using PyTorch.
- Explore geometric vision, calibration, localization, pose estimation, and 3D perception.
- Optimize models for low-latency inference and edge deployment.
- Build evaluation tools and support deployments on factory floors.
Requirements
- Currently studying computer science, robotics, electrical engineering, mechanical engineering, or a related field.
- Comfortable programming in Python with hands-on experience in computer vision, machine learning, or robotics.
- Hands-on experience with robotics or perception using real sensor or camera data.
- Curiosity about computer vision and physical AI, with the ability to learn quickly, build, debug, and take ownership.
- Experience with PyTorch, OpenCV, object detection, tracking, synthetic data, SLAM/VIO, 3D vision, CUDA/TensorRT, ROS2, or NVIDIA Jetson is a plus.
Benefits
- Visa sponsorship is available.
- Relocation support from Germany is supported.
- The role is on-site in San Francisco, California, United States.