About the position
Zoox is building advanced self-driving hardware and software solutions. To attain the utmost efficiency that the system demands, we need you - an expert that understands both compute hardware architecture as well as the algorithms and middleware that run on it. Your contributions will be instrumental in achieving optimal power levels while maximizing throughput. The Software Core Performance team’s mission is to analyze, optimize and provide guidance to the software and hardware teams in order to meet expected system performance targets. As a performance software engineer within the Software Core Performance team, you will instrument, monitor, analyze and optimize algorithms that are performance-critical. Your broad experience in developing, debugging, and optimizing software will enable AI software to run as efficiently as possible.
Responsibilities
- Build real-time instrumentation for performance monitoring (CPU, GPU, latency, memory) of the online system, and benchmarking frameworks to support offline performance evaluation.
- Build tools and scripts to evaluate and analyze performance at scale in CI, as well as on vehicle.
- Establish budgets for existing architectures, and provide data to define next-gen architectures.
- Analyze performance metrics in online software to identify resource utilization hotspots and root causes.
- Propose and co-implement actionable solutions with software component teams.
- Support AI software teams in squeezing out the maximum performance of their code, while minimizing resource contention.
- Work as part of the Core team to build a middleware framework that promotes efficient and performant code development.
- Work cross-functionally to ensure Zoox software runs optimally across the fleet.
Requirements
- BS in computer science or related field
- 6+ years of professional experience in software development or related fields
- Strong knowledge of C++ and experience in large code bases, proficiency with Python
- Familiar with CPU system architecture and OS fundamentals
- Comfortable in Linux development environments
- Good communication and organization skills, with a logical approach to problem solving, good time management, and task prioritization skills
- Solid understanding of the Linux operating system
Nice-to-haves
- Experience using various Linux performance monitor tools, such as perf, eBPF, Perfetto
- Experience with software & hardware benchmarking and Hardware-in-the-Loop (HIL) systems
- Proficiency with Git, Bazel, and CI pipelines
- Hands-on experience in development, debugging and profiling of complex multi-process real time systems like game engines or robot systems
- Acquainted with a firm/hard real-time operating system environment
- Proficiency with SQL and data platforms such as DataBricks, Looker, and ETL pipelines
- Experience debugging and optimizing GPU kernels using tools like Nsight Systems and Compute
Benefits
- Paid time off (e.g. sick leave, vacation, bereavement)
- Unpaid time off
- Zoox Stock Appreciation Rights
- Amazon RSUs
- Health insurance
- Long-term care insurance
- Long-term and short-term disability insurance
- Life insurance
