Software Engineer - Dexterous Manipulation

Apptronik
Austin, Mountain View
Workplace: OnsiteFull timeFunction: Software EngineeringExperience: 3+ yearsSkills: ["Code quality","Peer review","Collaboration","Documentation","Debugging"]

Build and productionize reinforcement learning and learning-based control for dexterous, multi-fingered robotic hands. Implement and tune real-time manipulation algorithms that grasp, perform in-hand manipulation, and integrate tactile feedback with low-latency execution. Translate RL/imitation and human-to-robot retargeting methods into production-grade C++/Python, develop sim-to-real pipelines in IsaacSim/MuJoCo/Drake, and deploy on physical robots while diagnosing sensor, latency, and calibration issues.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Apptronik
Apptronik
3 days ago

Software Engineer - Dexterous Manipulation

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 3 hours agoStatus: Live

Job Summary

Build and productionize reinforcement learning and learning-based control for dexterous, multi-fingered robotic hands. Implement and tune real-time manipulation algorithms that grasp, perform in-hand manipulation, and integrate tactile feedback with low-latency execution. Translate RL/imitation and human-to-robot retargeting methods into production-grade C++/Python, develop sim-to-real pipelines in IsaacSim/MuJoCo/Drake, and deploy on physical robots while diagnosing sensor, latency, and calibration issues.
Location: Austin, Mountain View
Workplace: Onsite
Employment Type: Full time
Job Function: Software Engineering
Seniority: Mid level

Key Responsibilities

  • •Implement and tune control algorithms for multi-fingered hands including grasping, in-hand manipulation, and tactile-feedback integration.
  • •Develop and maintain low-latency, reliable manipulation software inside the real-time controls stack.
  • •Translate RL/imitation and human-to-robot motion retargeting methods into production-grade C++/Python.
  • •Build and refine sim-to-real pipelines (hand models, domain randomization, validation) using IsaacSim, MuJoCo, and Drake.
  • •Deploy and debug manipulation on physical robots and diagnose sensor-noise, latency, and calibration issues while collaborating with hardware and systems teams.

Key Requirements

  • •Experience with dexterous manipulation for multi-fingered grasping, in-hand manipulation, and high-DOF hand control.
  • •Reinforcement learning for robotic control (reward design, training, debugging), ideally for dexterous manipulation (imitation learning and diffusion policies are a plus).
  • •Strong Python and working C++ experience for real-time robotic software.
  • •Hands-on experience training and validating policies in a physics simulator such as IsaacSim, MuJoCo, or Drake.
  • •Ability to get algorithms working on physical hardware, not simulation alone.
Experience:3+ yearsRobotics
Education:
Skills:Code qualityPeer reviewCollaborationDocumentationDebugging
Languages:En
Tech Stack:C++PythonIsaacSimMuJoCoDrakeReinforcement learningImitation learning

Company Brief

Apptronik
Develops advanced general-purpose robots and robotic limbs for industrial and commercial applications, combining hardware, actuators, and control software to enable versatile, human-scale robot solutions.
Industry: Robotics
Company Size: Medium (51 to 250 employees)
Growth: Growth Stage Startup
Funding: Series A
Headquarters: Austin, United States
Founded: 2018
WebsiteLinkedIn