GPU Software Engineer - High-Perf Compute & Profiling

Waymo; New York (NY); Hybrid; USD 204,000 - 259,000; Full time

Benefits offered by this job

Discretionary annual bonus

Equity incentive plan

Benefits for employees

Job summary

Waymo is seeking a GPU Software Engineer to advance the driverless platform. The role focuses on developing high-performance GPU primitives and abstractions, collaborating with the hardware team on SoC projects, and bringing up the GPU stack on new platforms.

You will also contribute to testing infrastructure to strengthen CI/CD and create profiling tools for performance analysis. The ideal candidate has a strong C++ background, 3+ years in the field, and deep knowledge of GPU hardware.

Qualifications

• Bachelor degree in EECS with 3+ years of relevant industry experience.
• Proven expertise in C++ programming.
• Experience with full-system simulation frameworks (SystemC, Gem5 or similar).
• Solid understanding of GPU hardware architecture.
• Proficiency with performance analysis tools and debuggers.

Responsibilities

• Develop high-performance GPU primitives and abstractions for scaling the accelerator codebase across GPU backends.
• Collaborate with internal hardware team and external partners on SoC projects focusing on the GPU portion.
• Manage bring-up, correctness, and performance of the onboard stack on new GPU platforms.
• Contribute to testing infrastructure to enhance CI/CD, detect bugs early, and generate automated alerts.
• Create profiler and debugger tools for new GPU platforms.

Skills

C++ programming

GPU architecture

Performance analysis

Debugging tools

GPU software stack

Education

Bachelor in EECS

Tools

SystemC

Gem5

Job description

Waymo is seeking a GPU Software Engineer to advance the driverless platform. The role focuses on developing high-performance GPU primitives and abstractions, collaborating with the hardware team on SoC projects, and bringing up the GPU stack on new platforms.

You will also contribute to testing infrastructure to strengthen CI/CD and create profiling tools for performance analysis. The ideal candidate has a strong C++ background, 3+ years in the field, and deep knowledge of GPU hardware.