Google is seeking a Staff Software Engineer specializing in TPU performance optimization, responsible for enhancing the efficiency of Tensor Processing Units and collaborating with product teams to solve performance challenges in machine learning. The ideal candidate should have a strong background in software engineering, machine learning, and performance analysis
Job Summary
This role focuses on squeezing maximum efficiency out of Google's production and research workloads using TPUs.
Engineers will collaborate across JAX and PyTorch to solve performance problems and onboard new ML models onto TPU hardware.
The position offers opportunities to drive breakthrough capabilities in AI infrastructure at unparalleled scale and velocity.
Matching Summary
Match Score: 85
Google is seeking a Staff Software Engineer specializing in TPU performance optimization, responsible for enhancing the efficiency of Tensor Processing Units and collaborating with product teams to solve performance challenges in machine learning. The ideal candidate should have a strong background in software engineering, machine learning, and performance analysis.
Salary
Base: $207,000-$300,000; Bonus/Equity: Included; Benefits: Not specified
Skills & Requirements
Must-have
Tensor Processing Unit (TPU) efficiency analysis
JAX and PyTorch framework expertise
Machine Learning training and serving benchmarks
Compiler and runtime optimization skills
Model co-design and quantization techniques
Nice-to-have
Experience with Gemini model architecture
Collaboration with product teams and researchers
Knowledge of large-scale distributed systems
Interest in hardware-software co-design
Passion for pushing technology boundaries
Key Requirements
Technical expertise in software engineering
Ability to manage project priorities and deadlines