5 months ago
Base Salary
$217k - $307k/yr
Responsibilities
- Build real-time CPU, GPU, latency, and memory instrumentation for performance monitoring.
- Develop offline benchmarking frameworks, tools, and scripts to evaluate and analyze performance at scale in CI and vehicles.
- Establish performance budgets for next-generation compute architectures.
- Analyze performance metrics to identify GPU hotspots and root causes.
- Propose and co-implement performance solutions with component teams.
- Help teams bring serial algorithms to the GPU to maximize compute utilization and reduce latency.
- Design a middleware framework that encourages efficient and performant CPU and GPU code development.
Requirements
- Bachelor of Science in computer science or a related field.
- At least 7 years of experience.
- Strong knowledge of CUDA and recent GPU microarchitectures such as Ampere and Blackwell.
- Experience debugging and optimizing GPU kernels with tools such as Nsight.
- Strong C++ knowledge and experience working in large codebases in Linux environments.
- Experience developing, debugging, and profiling complex multiprocess systems such as robotic systems or game engines.
- Bonus: GPU kernel development in real-time environments, PTX-level programming, CPU SIMD instructions including AVX intrinsics, and custom CUDA layers with TensorRT and XLA.
- Bonus: machine-learning model optimization, including post-training quantization and layer pruning, or GPU kernel tuning with OpenGL, CUDA, ROCm, or similar technologies.
- Bonus: proficiency with SQL, Databricks, Looker, or other business intelligence tools.
About 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.