The Software Engineer III position focuses on bridging the gap between machine learning workloads and custom Tensor Processing Unit (TPU) hardware. The role involves analyzing and optimizing distributed systems and compiler architectures to enhance performance across Google's AI infrastructure. Engineers will collaborate with various product areas, such as Google Cloud and YouTube, to onboard novel workloads and drive TPU adoption. This position requires versatility and leadership qualities as the team tackles new challenges in a fast-paced environment.

• 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.

• Develop and scale benchmarking and workload characterization strategies for TPU optimization.
• Drive full-stack hardware-software co-design to enhance ML accelerator architectures.
• Partner with product areas to efficiently scale key workload pipelines during TPU transitions.
• Build and upgrade compiler-aware simulator tools and performance-ladder pathways.
• Distill complex performance analyses into presentations for decision-making in leadership forums.

• Employees at Google are often offered benefits like comprehensive health insurance, 401(k) matching, and flexible work arrangements, among other benefits.