Software Engineer- Human Motion Data

Apptronik Inc.
Austin, United States of America
21 days ago

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Working hours
Regular working hours
Languages
English
Experience level
Junior

Job location

Remote
Austin, United States of America

Tech stack

Computer Animation
C++
Video Game Development
Game Engine
Python
Kinematics
Motion Capture
Motion Planning
Open Source Technology
Data Streaming
Reinforcement Learning
Data Processing
Delivery Pipeline
Information Technology
Unreal Engine
Data Pipelines
Data Generation

Job description

  • As a Software Engineer- Human Motion Data, you will leverage your background in robotics to build the crucial link between human-data and our reinforcement learning pipelines
  • This role is dedicated to architecting robust motion data pipelines-integrating diverse sources like mocap, game engines (like Unreal or Unity), teleoperation, and generative AI motion models to generate thousands of rich, physically accurate human motion trajectories
  • You will apply your deep expertise in kinematics and rigid body dynamics to translate raw human movement into actionable, dynamically feasible data for whole-body reinforcement learning
  • As a core member of the Motion Control and Planning team, you will work closely with Controls stakeholders to play a key role in maintaining a high-velocity, ego-free engineering culture while ensuring our humanoid robots move with unprecedented fluidity
  • Design, build, and maintain end-to-end motion data pipelines, integrating diverse sources such as motion capture (mocap), teleoperation, and synthetic generation using diffusion models, animation and gaming engines, to support humanoid robot development
  • Implement and optimize kinematic and dynamic retargeting pipelines to accurately map human demonstrations onto the robot's specific physical constraints, mass distributions, and joint limits
  • Develop tools and scripts to process and clean raw human demonstration data, and apply state-of-the-art retargeting libraries (e.g., GMR, Omni-retarget) to synthesize and filter new behaviors
  • Leverage game engines (Unreal Engine or Unity) and physics simulators to build simulated environments for procedural motion generation and data augmentation
  • Generate high-volume, high-quality trajectory datasets required for training whole-body reinforcement learning policies
  • Write robust, automated pipelines to streamline data flow between human demonstration sources, generative motion models, game engines, and the RL training infrastructure
  • Collaborate closely with the Reinforcement Learning and Controls teams to iterate on data requirements, understand failure modes, and ensure the generated trajectories are physically viable on hardware

Requirements

  • Prolonged periods of sitting at a desk and working on a computer
  • Must be able to lift 15 pounds at times
  • Vision to read printed materials and a computer screen
  • Hearing and speech to communicate, * Experience building or maintaining pipelines for spatial data, including motion capture, teleoperation tracking, or AI-driven motion generation
  • A results-oriented mindset with an eagerness to bridge the gap between human motion data and real-world robotic control algorithms
  • Strong theoretical and practical understanding of robot kinematics (FK/IK), coordinate transformations, and rigid body dynamics
  • High adaptability and a willingness to explore new tools, moving seamlessly across different layers of the robotics software stack as project needs evolve
  • Hands-on experience with state-of-the-art motion generation models and open-source retargeting libraries (e.g., GMR, Omni-retarget)
  • Proficiency in C++ is highly valued to help integrate with our broader robotics software stack
  • Hardware (HW) experience, particularly working directly with physical robotic platforms
  • Familiarity and hands-on experience utilizing 3D game engines (Unreal Engine or Unity) or advanced physics simulators for robotics data generation or simulation
  • A strong portfolio showcasing relevant robotics, sim-to-real, or motion generation projects
  • A BS or MS degree in Robotics, Mechanical Engineering, Computer Science, or a related highly technical field
  • 2+ years of industry or applied research experience in robotics, motion planning, sim-to-real pipelines, or technical animation data generation
  • A proven track record of processing large amounts of spatial or motion data to drive robotic or simulated systems

Apply for this position