Staff ML Accelerator Architect
About The Position
Waymo’s Compute Team develops the compute platform that runs the fully autonomous vehicle software stack. The team architects and creates high-performance custom silicon, develops system-level compute architectures, and collaborates across disciplines to optimize hardware and software performance. The Staff ML Accelerator Architect analyzes inference workloads and maps them efficiently to hardware, develops performance-modeling tools, and works with compiler and model developers to shape engineering tradeoffs and future model architectures. The role also architects, simulates, and designs hardware solutions for autonomous driving models and works with cross-functional engineering teams to realize architectural goals. The position reports to a Hardware Engineering Manager.
Requirements
- M.S. degree in Computer Science or Computer Engineering, or equivalent practical experience.
- 10+ years designing or architecting complex, high-performance architectures such as CPUs, GPUs, and/or ML accelerators, in industry or through doctoral research.
- 3+ years of experience with machine learning architectures, acceleration, and model optimization.
- Deep knowledge of ML model architectures and their compute characteristics, including bottlenecks and optimization opportunities.
- Strong programming and algorithmic problem-solving skills.
Nice To Haves
- Track record of analyzing workloads and architecting and delivering novel hardware and software solutions that substantially improve ML inference performance and efficiency.
- Proven ability to lead cross-functional teams to deliver an architectural vision.
- Familiarity with novel semiconductor technologies and their tradeoffs across the full hardware stack.
- Experience with microarchitecture design using SystemVerilog or HLS.
Responsibilities
- Analyze inference workloads, map them efficiently to hardware, propose hardware-friendly implementations, and project performance.
- Build scalable tools for performance modeling and evaluation of inference workloads.
- Work with compiler and model developers to influence engineering tradeoffs and future model architectures.
- Architect, simulate, and design hardware solutions for state-of-the-art autonomous driving models.
- Collaborate with cross-functional engineering teams to identify opportunities and realize architectural vision.
Benefits
- Discretionary annual bonus program
- Equity incentive plan
- Company benefits program
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
