Software Engineer, GPU
Waymo - Mountain View, CA
posted 4 days ago
About the position
Waymo is an autonomous driving technology company with the mission to be the most trusted driver. Since its start as the Google Self-Driving Car Project in 2009, Waymo has focused on building the Waymo Driver-The World's Most Experienced Driver-to improve access to mobility while saving thousands of lives now lost to traffic crashes. The Waymo Driver powers Waymo One, a fully autonomous ride-hailing service, and can also be applied to a range of vehicle platforms and product use cases. The Waymo Driver has provided over one million rider-only trips, enabled by its experience autonomously driving tens of millions of miles on public roads and tens of billions in simulation across 13+ U.S. states. Waymo's Compute Team is tasked with a critical and exciting mission: We deliver the compute platform responsible for running the completely autonomous vehicle's software stack. To achieve our mission, we architect and create high-performance custom silicon; we develop system-level compute architectures that push the boundaries of performance, power, and latency; and we collaborate with many other teammates to ensure we design and improve hardware and software for maximum performance. In this hybrid role, you will report to a Senior Staff Engineer.
Responsibilities
- Collaborate with application teams to understand how to map newly developed algorithms to GPU, allowing our cars to 'see' further, operate smarter, and react faster
- Build primitives and abstractions that allow for scaling our code-base to constantly evolving workloads and hardware
- Improve and add new compiler optimizations to promote producing optimized GPU assembly
- Analyze performance counters, GPU micro-architectural features, and algorithms to identify optimization opportunities
- Contribute to infrastructure that performs testing/static analysis to catch bugs early and create automated alerts to encourage following best GPU performance practices
- Co-design hardware features and evaluate trade-offs for future generations of our compute platform
Requirements
- 5+ years experience C++ programming skills
- Advanced degree in Computer Science, similar technical field of study, or equivalent practical experience
- 5+ years experience with GPU programming/optimization using CUDA or similar technologies
- 1+ years experience with GPU architecture/programming model
- 5+ years experience using performance analysis tools and debuggers
- 5+ years experience with parallel computing/programming
Nice-to-haves
- Experience with LLVM or SPIR-V open-source ecosystems or other compiler projects
- Operating systems or embedded software experience especially working on device drivers
- Experience with GPU optimization techniques: memory coalescing, register/shared memory tiling, pinned memory, and warp-level programming
- Experience with GPU libraries: CUB, CUTLASS, Thrust, or Eigen
- Research experience in parallel algorithms, compilers, or computer architecture
- Experience with graphics workloads and shader programming
- ML frameworks/compiler/library
Benefits
- Discretionary annual bonus program
- Equity incentive plan
- Generous Company benefits program
