Software Engineer, Apptronik

• Proficiency with at least one: C++/Python
• Experience developing controls for robotics applications
• Demonstrated competence with Git or version control of distributed systems
• Ability to work in Linux and write shell scripts in bash
• At least one year of real-time, hardware-in-the-loop software development
• Bachelor's or Master's degree in Computer Science, Engineering, or a related technical field
• Minimum of 2 years of professional, full-time experience building and maintaining reliable, scalable systems
• 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

Desirable

• We operate at the cutting edge of embodied AI, applying our expertise across the full robotics stack to solve some of society's most important problems. You will join a team dedicated to bringing Apollo to market at scale, tackling the complex challenges like safety, commercialization, and mass production to change the world for the better
• We are seeking a Dexterity Engineer to build the software bridge between human intent and robotic execution. In this role, you will be a core contributor to the teleop stack, ensuring our humanoid, Apollo, can perform complex, contact-rich tasks with high fidelity
• You will accelerate bimanual manipulation in at least two ways: Firstly, by developing the retargeting and haptic frameworks that make teleoperation intuitive for the operator
• Secondly, by building the simulation and testing infrastructure that validates end-effectors (COTS and internal designs) while ensuring every movement generates training-ready data for autonomy and reinforcement learning pipelines
• Generic Dexterity: Implement and refine retargeting algorithms (IK-based, optimization-based, or learned) that can map human hand kinematics to varied robotic configurations
• Signal Integrity: Address real-world teleop challenges: jitter reduction, handling sensor dropouts, and implementing safe fallback behaviors
• End-Effector Validation: Build automated workflows in MuJoCo/IsaacSim environments and real-robot testbeds to benchmark hand designs
• Performance Metrics: Develop automated scoring for dexterity to inform hardware and software design
• Sensory Loop: Integrate haptic gloves and tactile sensor data into the control loop to provide operators with force-reflective feedback
• Data Consumption: Ensure high-bandwidth tactile data is cleanly ingested into the robot state and logged for model training
• Fidelity Improvements: Research and implement control methodologies that improve the transparency of the teleoperation experience
• RL Scaffolding: Build the framework required to swap manual teleoperation with learned policies, enabling "shared autonomy" where the robot assists the operator in complex manipulations