Staff Software Engineer, ML Performance, GPU
Sunnyvale, CA, USA · Staff · Full-time · Posted Sep 23, 2026
- Experience
- 8 years in software development; 5 years with ML design and ML infrastructure. Preferred: 8 years with data structures and algorithms; 3 years in technical leadership; 3 years in a complex, matrixed organization.
- Education
- Bachelor’s degree or equivalent practical experience; preferred Master’s degree or PhD in Engineering, Computer Science, or a related technical field.
About the job
Optimize and benchmark GPU systems for Google’s machine-learning workloads, improving performance and efficiency across its infrastructure.
What you'll do
- Maintain LLM training and serving benchmarks to identify performance opportunities and guide XLA:GPU/Triton performance and XLA releases.
- Partner with product teams to onboard, optimize, and scale LLMs and other ML models on GPU hardware.
- Run architecture simulations, benchmarking, and roofline analyses; use fleet-wide metrics and profiling to identify bottlenecks and implement optimizations.
What they ask for
- Bachelor’s degree or equivalent practical experience.
- 8 years of experience in software development.
- 5 years of experience with ML design and ML infrastructure.
- Experience with modern GPU architectures and low-level GPU programming, performance engineering, and modern LLM deployment on AI accelerators.
Nice to have
- Master’s degree or PhD in Engineering, Computer Science, or a related technical field.
- Experience with data structures and algorithms, technical leadership, complex matrixed organizations, hardware-aware algorithm design, or compiler stacks such as OpenXLA.
Pay
$207000 - $300000 USD base salary
Tools
XLA:GPU, Triton, TRT-LLM, vLLM, SGLang, CUDA, CUTLASS, OpenXLA
AI focus
Improve GPU performance, deployment, and training efficiency for LLMs and other machine-learning workloads.
Model-written summary · the original posting is authoritative