Software Engineer – System Identification

JOB SUMMARY

This role is responsible for essential system identification software and processes that enable precise, high-performance control of Apptronik robots. They will ensure that models employed by the robot abstraction API used for planning and controls matches both simulation and hardware across a fleet of dynamic humanoid platforms, for everything from actuation modeling to whole-body dynamical system characterization. Through experimental procedure, knowledge of first principles, and modular software design, they will ensure that both model-based and model-free behaviors developed in simulation transfer seamlessly to hardware deployments at scale. They will collaborate extensively with hardware, software, and test teams, and implement extensive system testing to validate this performance. Their expertise in software design, system identification, and controls fundamentals will drive the innovation that brings our robots to life.

ESSENTIAL DUTIES AND RESPONSIBILITIES or KEY ACCOUNTABILITIES

• Design software and processes to ensure that measured robot hardware quantities, simulation plants, and control models align
• Design and execute hardware experiments to accurately characterize kinematic and dynamic systems
• Collaborate with simulation engineers to ensure that closed-loop simulation behaviors match reality
• Provide software support of system testbeds used for characterization and QA
• Implement software to ensure that models remain accurate across a fleet of robots deployed in the field over their lifecycles
• Employ data-driven methods to improve robustness to unmodeled system dynamics
• Implement rigorous unit and integration testing of system identification and modeling
• Maintain comprehensive and accurate architecture and design documentation
• Deliver reliable software through code reviews, continuous integration, and automated testing
• Represent and lead with Apptronik's core values: CHIPC - Curiosity, Humility, Integrity, Passion, Creativity

SKILLS AND REQUIREMENTS

• Proven experience conducting system identification for dynamic robots (legged preferred)
• Strong modern C++ and Object-Oriented Programming skills, including inheritance, templates, and design patterns
• Strong theoretical understanding of robotics fundamentals such as kinematics, dynamics, controls, and estimation
• Familiarity with common robotic testbed equipment such as motion capture, force plates and dynamometers
• Experience with common robotics and controls packages such as ROS, URDF, MuJoCo, and Eigen
• Prior experience tuning robotic simulators for improved fidelity
• Experience developing Linux applications
• Software testing methodologies (unit & integration testing)
• Git or similar version control systems
• Strong technical communication skills
• Familiarity with Agile process (sprint planning, design reviews, & code reviews)

EDUCATION and/or EXPERIENCE

• BS, MS or PhD degree (or equivalent) in Computer Science, Computer Engineering, Electrical Engineering, Mechanical Engineering, Robotics, or related field
• Proven industry experience integrating robot hardware and software (5 years, or 2 with grad degree)

PHYSICAL 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