Staff Software Engineer, Simulation
Hybrid- Any Office (Fremont, CA, Salem, OR, or Pittsburgh, PA) · Hybrid- Fremont, CA +1
$181k–$282k a yearRemoteLead & above
Job details
- Hybrid- Any Office (Fremont, CA, Salem, OR, or Pittsburgh, PA) · Hybrid- Fremont, CA +1
- Remote
- $181k–$282k a year (posted by the company)
- Full-time · Lead & above
- Software Engineering
- Posted 6 days ago
Sponsorship history
Past filings by Agility Robotics in public government data. Not a promise of sponsorship.
- Filed 5 H-1B LCAs in FY2025
Full job description
Agility’s commercially deployed humanoids operate alongside teams in warehouses, manufacturing facilities, and distribution centers—tackling physically demanding and repetitive tasks while enabling workers to focus on higher-value work. With industry-leading safety standards and years of proven deployment data, we're pioneering a new era of automation that enhances human potential.
About the Role
Agility Robotics is building the world's most capable humanoid robots, and we're looking for a senior engineer to help build the simulation software at the foundation of it. You'll be one of the first engineers on a new Simulation team, working alongside the team's leader to shape its technical direction, architecture, and infrastructure from the ground up.
You'll work at the intersection of physics simulation, hardware abstraction, and robot learning, partnering closely with our AI, Controls, Perception Sensors, Hardware Interfaces, Quality Assurance, and Systems Integration teams to build simulation environments that accelerate development, provide insights into the system, and enable rapid validation.
This role can be hybrid (3 days on-site per week at one of our offices) or fully remote, with up to one week of travel per month.
About the Work
- Design, build, and own core components of Agility's simulation stack as a founding member of the Simulation team
- Develop the hardware abstraction layer connecting sensors, actuators, and physical hardware to simulation environments
- Develop and integrate physics simulators using open-source frameworks, with a focus on fidelity, performance, and sim2real transfer
- Build and maintain high-fidelity models of new and existing robots, including kinematics, dynamics, mass properties, actuator and transmission models, and contact behavior
- Build robust automated pipelines from hardware design, to robot model, to system-identification, allowing new hardware designs to rapidly be used in sim
- Use simulation to train RL policies, generate synthetic data and evaluate end-to-end models
- Use simulation to evaluate proposed robot designs, running trade studies on things like actuator sizing, link geometry, range of motion, and sensor placement, and turn the results into clear recommendations for design teams
- Validate models against physical hardware data and improve fidelity where simulation and reality diverge
- Set a high bar for engineering practices through design reviews, code reviews, and mentorship of other engineers as the team grows
