Apptronik
Software Engineer - Dexterous Manipulation
Austin, TX +1 · Full-time · Posted Aug 5, 2026
- Location
- Austin, TX
Sunnyvale, California, United States - Experience
- 3+ years of relevant experience in robotic manipulation or complex motion control; recent PhD graduates considered.
- Education
- BS/MS/PhD in Robotics, Computer Science, Electrical Engineering, or a related field.
About the job
Implements, tunes, and deploys learning-based control for dexterous robotic hands in simulation and on physical robots.
What you'll do
- Develop control and manipulation software for multi-fingered hands, including grasping and tactile-feedback integration.
- Build sim-to-real pipelines and deploy, debug, and validate manipulation capabilities on physical robots.
- Translate reinforcement- and imitation-learning methods and motion-retargeting approaches into production software.
What they ask for
- Experience with dexterous manipulation, including multi-fingered grasping and high-DOF hand control.
- Reinforcement learning for robotic control, including reward design, training, and debugging.
- Strong Python, working C++, and hands-on policy training and validation in a physics simulator.
- Robotics fundamentals and demonstrated deployment of algorithms on physical hardware.
Nice to have
- Imitation learning and diffusion policies.
- Teleoperation and retargeting, tactile sensing, computer vision, or end-effector bring-up and calibration.
Tools
Python, C++, IsaacSim, MuJoCo, Drake
AI focus
Applies reinforcement learning and related learning-based control, including imitation learning, to dexterous robotic manipulation.
Model-written summary · the original posting is authoritative