Real-Time GPU Performance Engineer (C++/CUDA)

Front Door Defense; Foster City (CA); On-site; USD 168,000 - 239,000; Full time

Job summary

Zoox is seeking a Software Engineer - C++ GPU Performance to build a high‑performance GPU‑accelerated software stack for autonomous vehicle perception workloads. You will instrument, monitor, and optimize GPU‑based algorithms, helping influence compute platform design and resource allocation for real‑world autonomous systems.

You will work on real‑time instrumentation, offline benchmarking, and identifying GPU hotspots.

Qualifications

• BS in computer science or related field.
• 3+ years of experience in software development.
• Strong knowledge of CUDA and GPU debugging with Nsight.

Responsibilities

• Build real-time instrumentation for performance monitoring (CPU, GPU, latency, memory).
• Analyze performance metrics to identify GPU hotspots and root causes.
• Help bring serial algorithms to the GPU to maximize compute utilization and reduce latency.
• Contribute to designing a middleware framework to promote efficient code on CPU and GPU.

Skills

C++

CUDA

Linux development

Performance analysis

Education

BS in computer science or related field

Tools

Nsight

TensorRT

XLA

Job description

Zoox is seeking a Software Engineer - C++ GPU Performance to build a high‑performance GPU‑accelerated software stack for autonomous vehicle perception workloads. You will instrument, monitor, and optimize GPU‑based algorithms, helping influence compute platform design and resource allocation for real‑world autonomous systems.

You will work on real‑time instrumentation, offline benchmarking, and identifying GPU hotspots.