GPU Performance Engineer

Gpu Performance Engineer

Google's software engineers develop the next-generation technologies that change how billions of users connect, explore, and interact with information and one another. Our products need to handle information at massive scale, and extend well beyond web search. We're looking for engineers who bring fresh ideas from all areas, including information retrieval, distributed computing, large-scale system design, networking and data storage, security, artificial intelligence, natural language processing, UI design and mobile; the list goes on and is growing every day. As a software engineer, you will work on a specific project critical to Google's needs with opportunities to switch teams and projects as you and our fast-paced business grow and evolve.

Gpu Performance Team is responsible for optimizing, modeling and evaluating Gpu systems for comparative analysis and benchmarking for Google's internal Machine Learning (Ml) workloads. We strive for extracting maximum efficiency in Google's growing Gpu fleet. The team identifies performance opportunities in Google production and research Ml workloads and lands improvements that impact the entire fleet. The team also evaluates current and future Ml workloads through simulations and comparative benchmarks to guide decision making for the Cloud hardware teams.

Responsibilities:

• Build improvements for the latest generation of Graphics Processing Units (Gpus) that power Google's most critical products and services, impacting billions of users worldwide.
• Identify performance bottlenecks and drive improvements across the breadth and depth of Google's Gpu software stack from Ml compiler cost model design (Openxla, Triton, Mlir) to optimizing performance Gpu kernels (Pallas Mosaic, Cute) to cross node model serving configurations (e.g., disaggregated serving, paged attention).
• Influence the technical direction of the Gpu software ecosystem at Google by collaborating with Modeling, Xla:Gpu, Deepmind and Performance Tooling teams. Influence the deployment of Google's Gpu fleet by working with various product teams across Google.