Robotics Data Engineer, Human Motion Data

Our exoskeletons are deployed in real warehouses and production plants and capture what video-only datasets miss: body pose, joint kinematics, interaction torques, vision, and hand forces from real human work. Your job is to build pipelines that turn raw exoskeleton recordings into clean, robot-ready training data, including retargeting human motion to the humanoid’s body. So in the future human workers and robots can work side by side on the same tasks.

What you’ll do

• Define signals, formats and metadata for our next exoskeleton generation
• Build pipelines from devices in the field to storage, incl. sync, calibration and versioning
• Build tooling for quality checks, segmentation, annotation and dataset export
• Implement anonymization, consent handling and GDPR-compliant data flows
• Set up storage and compute together with our robot learning team

• Degree in CS, data engineering, robotics or similar, with hands-on pipeline experience
• Strong Python and SQL; pipeline tools (e.g. Airflow, Ray or Spark)
• Time-series or multimodal sensor data, incl. synchronization and calibration
• Cloud (AWS, GCP or Azure), Docker and Linux
• Solid understanding of GDPR and personal data

Nice to have

• ROS 2, MCAP or LeRobot; annotation or dataset management tools
• Kubernetes or SLURM; basic ML training workflows in PyTorch, EASE is an equal opportunity employer. Applicants need a valid work permit for the EU.

• Data almost nobody else has, from a product in series deployment
• The chance to build our data platform from the ground up
• Competitive salary plus virtual stock options (VSOP)
• Flexible hours, hybrid in Munich, mobility benefit
• Budget for conferences and further training

EASE builds active, AI-powered exoskeletons for logistics and manufacturing. As a TUM spin-off, we have just started series production, and paid pilots with well-known logistics companies have led to first purchase orders. Our goal: move every parcel better, manually with an exoskeleton or autonomously with a robot., EASE

Gardening and landscaping