Staff Software Engineer – Humanoid Controls

JOB SUMMARY 

This person has proven hands-on capabilities developing planning and control algorithms for humanoid robots as well as a strong theoretical background in both model-based and model-free methods.  They will help drive the technical strategy for humanoid controls, implement key state-of-the-art locomotion and manipulation algorithms (in either C++ or Python) in simulation and on hardware, and provide technical leadership required to deliver performant, reliable robot behaviors. They will also implement rigorous unit and integration tests of these and existing algorithms, quantify their performance, and develop new features to drive continuous improvement. They will work as part of a team focused on developing a dynamic humanoid robot robust to unstructured environments and able to perform useful work in real customer applications.  

ESSENTIAL DUTIES AND RESPONSIBILITIES or KEY ACCOUNTABILITIES:

• Deliver state-of-the-art planning and controls algorithms on high performance humanoid robot hardware 
• Characterize and improve the quality of robot locomotion and manipulation
• Contribute to the development of technical strategy and roadmaps for humanoid controls
• Collaborate with Autonomy, Perception, and other teams to enable a broad range of robot behaviors
• Implement rigorous unit and integration testing of control algorithms and implementation
• Maintain comprehensive and accurate technical 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 implementing state-of-the-art controls of dynamic legged robots on hardware. Model-free (e.g. reinforcement learning) required, model-based (e.g. MPC) nice-to-have
• Strong understanding of robotics fundamentals such as kinematics, dynamics, controls, estimation, and perception
• Strong understanding of reinforcement learning fundamentals such as policy optimization, value-based methods, deep learning architectures, and their applications to humanoid robotics
• Proficient in object-oriented programming in a Linux development environment, with prior demonstrated ability to:  
• Translate complex theoretical concepts and algorithms into practical, efficient, and bug-free software implementations
• Perform on a fast-paced software development team using standard CI tools like Git while following rigorous documentation and testing standards  
• Python required, C++ nice-to-have
• Familiarity with simulation tools and physics engines (e.g., Mujoco, Isaac Lab).
• Strong technical communication skills 

EDUCATION and/or EXPERIENCE:

• MS or PhD degree (or equivalent) in Computer Science, Computer Engineering, Electrical Engineering, Mechanical Engineering, Robotics, or related field 
• Proven industry experience implementing high-performance legged controls (7 years, or 5 with directly relevant PhD) 

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

Perks

• Competitive base w/ benefits
• Highly motivating equity incentive package
• Hybrid work schedule (i.e. work from home in some capacity) 
• Flexible work hours policy
• Professional development opportunities
• Open/Flexible Paid Time Off (PTO)
• 401k plan