TPU Performance Engineer — HW/SW Codesign

Google Inc.; Sunnyvale (CA); On-site; USD 147,000 - 210,000; Full time

Job summary

Google is hiring a Software Engineer III for TPU Performance, Hardware and Software Codesign in Sunnyvale, CA. You will bridge ML workloads with TPU hardware, optimizing distributed systems and compiler architectures to improve TPU adoption and performance.

Collaborating with Google Cloud and YouTube teams, you will deliver scalable performance analyses, guide roadmap decisions, and influence the next generation of ML infrastructure. Strong programming and systems knowledge are required.

Qualifications

• Bachelor’s degree or equivalent practical experience.
• 2 years of experience with software development in one or more programming languages.
• 2 years of experience with performance, large-scale systems data analysis, visualization tools, or debugging.
• 2 years of experience with computer architecture, performance analysis, and performance modeling.
• Master's degree or PhD in Computer Science or related fields.

Responsibilities

• Develop and scale benchmarking and workload characterization strategies to enable fast grounding-to-silicon, root-cause performance analysis, and TPU mapping optimization.
• Drive full-stack hardware-software co-design to optimize current and future ML accelerator architectures for business-critical production models (e.g., LLMs and embedding models).
• Partner with Product Areas to scale key workload pipelines efficiently during TPU Pilot and GA transitions.
• Build and upgrade compiler-aware simulator tools, hardware cost-models, and performance-ladder pathways to baseline and project physical silicon capabilities.
• Distill complex performance analyses and hardware trade-offs into presentations to guide TPU roadmap decision-making in core leadership forums.

Skills

Software development

Performance analysis

Computer architecture

Visualization tools

Education

Bachelor’s degree or equivalent practical experience

Master’s degree or PhD preferred

Tools

XLA

TPU

Job description

Google is hiring a Software Engineer III for TPU Performance, Hardware and Software Codesign in Sunnyvale, CA. You will bridge ML workloads with TPU hardware, optimizing distributed systems and compiler architectures to improve TPU adoption and performance.

Collaborating with Google Cloud and YouTube teams, you will deliver scalable performance analyses, guide roadmap decisions, and influence the next generation of ML infrastructure. Strong programming and systems knowledge are required.