Robotics Software Engineer, Simulation San Francisco, CA · On-site

Weave was founded to build the robots we’d want to have in our own home. We believe the next generation of robotics will transform everyday life by enabling people to do more and to reclaim time to spend on what’s important.

We also believe robots are in a sense like any other product: to matter, they have to ship. Our robots are already operating in real homes and businesses, giving us the opportunity to rapidly improve from real-world experience. With a growing team, strong customer demand, and capital for expansion, we’re entering an exciting stage of growth—and we’re looking for people with exceptional talent and standards to help bring home robotics to millions of households.

Simulation is essential at nearly every level of our stack: validating kinematics, iterating on controls research, stress testing new mechanisms, trying new manipulation strategies, and proving out the robustness of our planning and SLAM stack. How fast all of those loops run, and how honestly simulation predicts hardware, is decided by the software this role owns.

You’ll build our simulation platform as a core product with customers across the company: the environments, the models, and the fidelity work that makes "it works in sim" mean something.

Own the sim stack. Isaac Lab, Mu Jo Co , or equivalent: environments, assets, robot models, and the tooling around them, kept current with every hardware revision that ships.

Model what matters. Actuator models (torque limits, friction, thermal behavior) and contact parameters identified against dyno and robot data with the motor control and controls teams; you know which fidelity buys transfer and which buys nothing.

Make training fast. GPU-parallel environment throughput, profiling, and the determinism and seeding discipline that make experiments reproducible.

Serve sim-to-real. System-identification pipelines, domain randomization tooling, and honest side-by-side measurement of where simulation and hardware disagree.

Catch regressions before hardware does. Scenario and regression suites that run controllers and policies in sim on every change.

Keep assets true. The pipeline from CAD and robot description to sim-ready assets, so simulation tracks the robot we actually build.

3+ years building simulation or robotics software, with deep hands-on experience in at least one modern robotics simulator (Isaac Lab/Isaac Sim, Mu Jo Co , Drake, or similar).

Physics literacy: knowledge of contact solvers, fabrics solvers, integrators, and friction models, including the specific limits of each.

Strong Python and C++ and the engineering habits to make a research-facing platform reliable.

Performance skill: you’ve profiled and scaled GPU-parallel workloads and treat environment throughput as a feature.

Respect for the real robot. You’ve compared sim to hardware honestly and changed the sim.

RL training experience: you’ve been the customer of a sim stack and remember what hurt.

Rendering or synthetic-data experience for perception training.

USD or asset-pipeline experience at scale.