3 hours ago
San Mateo, CA, USAMid Level
H1B sponsor
Responsibilities
- Maintain and iterate on embodied AI data collection systems, including multimodal sensor synchronization, data pipelines, storage, and annotation toolchains.
- Integrate, calibrate, and routinely maintain ZED, RealSense, and industrial camera systems and robotic arms.
- Deploy imitation learning, DAgger, and teleoperation algorithms end to end on robots, including data collection, model training, inference deployment, and on-site tuning.
- Develop data quality assurance mechanisms for timestamp synchronization, coordinate frame alignment, anomaly detection, and collection-failure diagnosis and repair.
- Define and maintain robot-side software interfaces and communication protocols using ROS2 and DDS.
- Support on-site data collection and algorithm deployment, document technical processes, and improve collection efficiency and model performance.
Requirements
- Bachelor's degree or above in Robotics, Automation, Computer Science, or a related field, with 3+ years of relevant experience.
- Strong Python and C++ programming skills and familiarity with system-level development and debugging in Linux environments.
- Expertise in ROS and ROS2, including Topics, Services, Actions, parameter management, and the Launch system.
- Experience integrating and calibrating camera systems, including intrinsics, extrinsics, multi-camera synchronization, and ZED, RealSense, or similar RGB-D and stereo cameras.
- Experience with robotic arm control interfaces such as UR, Franka, xArm, or in-house arms, including driver layers, kinematics, and high-level APIs.
- Familiarity with CAN, EtherCAT, RS485, USB, IMUs, depth cameras, and force/torque sensors.
- Experience training and deploying deep learning models, with proficiency in PyTorch and the full workflow from data preparation and training through inference and deployment.
- Proficiency using AI coding tools such as Claude Code and Cursor in daily development.
- Preferred experience with embodied AI or imitation-learning data collection and deployment, Diffusion Policy, ACT, VLA, SVO2, ROS bag, HDF5, LeRobot datasets, camera hardware triggering, PTP synchronization, VIVE Trackers, OptiTrack, motion-capture gloves, video codecs, and Jetson hardware codecs.
- Fluent written English and ability to read research papers and open-source project documentation.
Benefits
- Competitive compensation package.
- Flat, open engineering culture with involvement in technical decisions.
- Well-equipped hardware lab resources for rapid prototyping and iteration.
- Flexible work arrangements and ongoing learning and internal tech-sharing programs.
- End-to-end ownership and autonomy across data collection, algorithms, and deployment.
Categories
ML EngineeringRobotics
About XDOF
XDOF builds data, software, and operational infrastructure for robotics companies and research labs developing general‑purpose robots and physical AI. It provides scalable data collection, tooling, hardware integration, and services that help teams train, deploy, and manage autonomy systems. The company is privately held and positions itself as an infrastructure partner to frontier labs and robotics builders, with a team drawing experience from Tesla, Google, Applied Intuition, BAIR, Covariant, and Gatik.