Staff Software Engineer, ML Performance and Accelerators

Requirements

• Bachelor’s degree or equivalent practical experience.
• 8 years of experience with software development in C++ and Python.
• 5 years of experience with ML infrastructure, including performance analysis and optimization of distributed systems.
• 5 years of experience modeling or evaluating performance for large-scale machine learning workloads (e.g., Large Language Models).
• Experience with hardware-software co-design, ML accelerators, and low-level software stacks such as compilers.

Nice to Haves

• Master’s degree or PhD in Computer Science or a related technical field.
• Experience in the self-driving or automotive industry building large-scale distributed infrastructure and networking systems.
• Experience in building high-fidelity HW/SW simulators (C++/Python) and large-scale performance tooling for ML accelerators.
• Understanding of large-scale AI workloads (LLMs), agentic systems, and performance/cost analysis for hyperscale systems.
• Knowledge of compiler flows, hardware-software co-design, and translating frameworks like PyTorch or JAX to hardware instructions.
• Ability to collaborate cross-functionally to translate performance insights into actionable hardware roadmap improvements.

What You'll Be Doing

• Define, implement, and validate performance and cost metrics for current and future workloads; analyze systems from chip to network to identify bottlenecks and optimize performance.
• Co-design hardware and system features for evolving workloads, identifying future hardware and software opportunities to improve ML performance.
• Build, maintain, and validate high-fidelity Hardware and Software (HW/SW) simulators in C++ and Python.
• Develop and maintain comprehensive benchmarks and propose changes to enhance overall infrastructure efficiency and end-to-end results.
• Collaborate with computer architects, hardware designers, and ML model owners to communicate findings and turn analysis results into actionable improvements.

Perks and Benefits

• Google's software engineers work on next-generation technologies that change how billions of users connect, explore, and interact with information.
• Opportunities to work on critical projects with the chance to switch teams and projects as the business evolves.
• Versatile role with leadership opportunities.
• A dynamic environment pushing technology forward.