Machine Learning / Computer Vision Engineer

Responsibilities

• Build computer vision and visual representation learning pipelines for robotic manipulation, including RGB, RGB-D, depth, segmentation, pose, keypoint, and object-centric representations.
• Develop visual models that support reinforcement learning and imitation learning policies, including end-to-end visuomotor policies that map visual observations to robot actions.
• Improve our data pipeline for vision-based manipulation policies through domain randomization, photorealistic rendering, synthetic data generation, sensor noise modeling, and real-world fine-tuning.
• Design and train perception models that are robust to lighting changes, camera viewpoint shifts, texture variation, clutter, occlusion, object instance variation, and imperfect calibration.
• Evaluate learned visual representations and policies on real robotic manipulation tasks, identify failure modes, and iterate on models, data, and training procedures.
• Collaborate with robotics, robot learning, and simulation engineers to define the perception strategy for robotic manipulation.
• Set up, calibrate, and evaluate camera and depth sensing systems when needed, with an emphasis on how sensor choices affect learned policies and real-world robustness.

Minimum Qualifications

• Ph.D. in computer vision or 3+ years of experience working on a computer vision product.
• Strong background in machine learning for computer vision, especially deep learning-based visual perception.
• Experience training modern computer vision models in Jax, PyTorch or similar frameworks.
• Practical experience with visual representation learning, object detection, segmentation, pose estimation, depth estimation, tracking, or 3D perception.
• Strong Python programming skills
• Ability to move fluidly between research code and production-quality systems.
• Strong understanding of how data distribution, sensor noise, calibration, lighting, and scene variation affect model performance.

Preferred Qualifications

• Experience training policies from visual observations, including RGB, RGB-D, point clouds, object-centric representations, or learned latent representations.
• Experience with domain randomization, synthetic data generation, differentiable rendering, neural rendering, or photorealistic simulation.
• Experience with robotics simulators or synthetic data tools such as Isaac Sim, MuJoCo or similar environments.
• Familiarity with robot learning methods such as reinforcement learning, behavior cloning, diffusion policies, offline RL, or learning from demonstrations.
• Experience with real robot deployment, including camera calibration, hand-eye calibration, depth sensors, ROS/ROS2, or robot data collection pipelines.
• First-author publications in top computer vision, robotics, or machine learning venues such as CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML, RSS, CoRL, ICRA, or IROS.