Senior/Staff Software Engineer - C++ Simulation Platform

Zoox•Foster City, CA

1d•$245,000 - $305,000

About The Position

Simulation is essential for Zoox to rapidly iterate on our driving software and hardware, and to validate our safety before we drive in the real world. We create virtual worlds to challenge our robots, from real-world data, entirely novel scenarios, or a combination of both. Our simulations need to run at a huge scale to cover everything that might happen, and to help prove our driving to be safe. As a Simulation C++ Platform Engineer, you will build the architecture on which simulation relies, with a specific focus on our high-speed, GPU-based simulation framework. Our success depends on you building solid, deterministic foundations that enable rapid scaling of our ML training pipelines and ensure seamless integration across the entire self-driving software stack.

Requirements

Bachelor's or Master's in computer science, mathematics, physics, or a related field.
6+ years of experience in a related software engineering field.
Deep expertise in modern C++ software development and system architecture.
System design experience, specifically building deterministic software in distributed computer clusters.
Strong understanding of execution frameworks, latency optimization, and memory management.

Nice To Haves

Familiarity with spatial reinforcement learning or modern ML frameworks (e.g., JAX, PyTorch).
Automotive, robotics, or AAA video game industry experience.
Familiarity with PubSub systems.

Responsibilities

Architect and support foundational simulation software, focusing heavily on execution frameworks and latency optimization.
Own the interface between our high-speed GPU simulation framework and the rest of the self-driving software stack.
Improve simulation performance directly and build self-serve systems that accelerate the work of autonomy and ML engineers.
Ensure the deterministic execution, precise timing, and reproducibility of all simulation results at scale.
Collaborate closely with ML experts to bridge fidelity gaps between fast simulations and high-fidelity 3D environments.