Staff TPU Performance Engineer: Optimize Large-Scale ML

Google; Kirkland (WA); On-site; USD 207,000 - 300,000; Full time

Benefits offered by this job

Health insurance

Dental, Vision, Life, Disability

401(k) with company match

Paid Time Off 20 days per year

Sick Time 40 hours/year (Seattle 69)

Maternity Leave 28-30 weeks

Baby Bonding Leave 18 weeks

Holidays 13 days

Job summary

Google is seeking software engineers to advance next-generation technologies that scale to billions of users. You will work on projects across JAX and PyTorch, focusing on TPU fleet efficiency and performance optimization, collaborating with product teams and researchers to solve complex ML deployment challenges.

The role emphasizes designing, developing, testing, deploying, and maintaining software solutions with opportunities to switch teams as Google grows.

Qualifications

• Bachelor's degree or equivalent practical experience.
• 8 years of experience in software development.
• 5 years of experience testing, and launching software products.
• 5 years of experience with performance, large-scale systems data analysis, visualization tools, or debugging.
• 3 years of experience with software design and architecture.
• Experience with ML performance analysis and benchmarking.

Responsibilities

• Learn more about benefits at Google .
• Focus on TPU fleet efficiency analysis and performance optimization, while identifying and maintaining ML training and serving benchmarks.
• Use benchmarks to identify performance opportunities and drive out-of-the-box performance by improving the compiler, runtime, etc. in collaboration with partner teams.
• Collaborate with Google product teams and researchers to solve performance problems, such as onboarding new ML models and products onto new TPU hardware to enable larger models to train efficiently at a very large scale.
• Analyze performance and efficiency metrics to identify bottlenecks, design, and implement solutions at Google fleet-wide scale.
• Explore model and data efficiency techniques i.e., model co-design, quantization, and sparsity.

Skills

ML performance analysis

Performance analysis

Debugging

Education

Bachelor's degree or equivalent practical experience

Master’s degree or PhD in Engineering, CS, or related field

Tools

OpenXLA

NVIDIA/AMD architectures

Job description

Google is seeking software engineers to advance next-generation technologies that scale to billions of users. You will work on projects across JAX and PyTorch, focusing on TPU fleet efficiency and performance optimization, collaborating with product teams and researchers to solve complex ML deployment challenges.

The role emphasizes designing, developing, testing, deploying, and maintaining software solutions with opportunities to switch teams as Google grows.