Architect onboard GPU execution for perception, prediction, and planning workloads, including scheduling, multi-tenancy, isolation, and deterministic latency.
Own the multi-year roadmap for reducing onboard compute per vehicle through latency and utilization budgets and optimization across models, runtimes, middleware, and hardware.
Guide GPU and SoC selection, compute topology, memory budgets, workload modeling, and next-generation platform bring-up.
Set direction for inference performance using TensorRT, CUDA, custom kernels, quantization, and collaboration with ML teams.
Lead and mentor senior engineers, align directors and VPs through design documentation, and resolve performance, safety, and schedule tradeoffs.
Requirements
Bachelor’s degree in computer science, computer engineering, or a related field.
10+ years of systems or performance engineering experience, including leading GPU or accelerator architecture across teams or building and leading a performance engineering team.
Deep expertise in the CUDA programming model, runtime, and driver, as well as NVIDIA GPU microarchitecture from Ampere through Blackwell, including Orin and Thor.
Proven record of setting technical direction and delivering measurable end-to-end improvements in latency, utilization, or cost.
Working knowledge of ML inference optimization, including TensorRT, XLA, Triton, quantization, and fusion.
Expert C++ development on Linux.
Bonus experience with real-time or embedded GPU systems, production GPU multi-tenancy, GPU platform bring-up, benchmarking methodology, or workload modeling and projections.
Salary: $339k - $455k/yr
Zoox
Zoox is transforming mobility-as-a-service by developing a fully autonomous, purpose-built fleet designed for AI to drive and humans to enjoy.