Senior GPU Performance Software Engineer, AI/ML

About the job

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. We need our engineers to be versatile, display leadership qualities and be enthusiastic to take on new problems across the full-stack as we continue to push technology forward.

The Machine Learning (ML)-Omega team is responsible for optimizing, modeling, and evaluating GPU systems for comparative analysis and benchmarking for Google’s internal ML workloads. We strive for extracting maximum efficiency in Google’s GPU fleet, evaluating current and future ML workloads to guide decision-making for the Cloud hardware teams.

Responsibilities

• Identify and maintain Large Language Model (LLM) training and serving benchmarks that are representative of Google production, industry, and the ML community, using them to identify performance opportunities, drive Accelerated Linear Algebra (XLA) Graphics Processing Unit (GPU)/Triton performance, and guide XLA releases.
• Engage with Google product teams like DeepMind to solve their ML model performance problems, such as onboarding new LLM models and products on GPU hardware and enabling LLMs to train and serve efficiently at a very large scale.
• Run architecture-level simulations on GPU designs and perform roofline analysis to guide internal teams.
• Analyze performance and efficiency metrics to identify bottlenecks, as well as design and implement solutions at Google fleet-wide scale.
• Run performance benchmarks on GPU hardware using internal and external tools.

Qualifications

Minimum qualifications:

• Bachelor’s degree or equivalent practical experience.
• 5 years of experience with software development in one or more programming languages.
• 5 years of experience with data structures and algorithms.
• 5 years of experience with machine learning algorithms and tools, artificial intelligence, deep learning, or natural language processing.
• 3 years of experience testing, maintaining, or launching software products, and 1 year of experience with software design and architecture.

Preferred qualifications:

• Master's degree or PhD in Computer Science or a related technical field.
• 1 year of experience in a technical leadership role.
• Experience with hardware/compiler co-design or high-performance computing (HPC).
• Experience in performance analysis and debugging, improving performance of single-node or multi-node (distributed) systems.
• Experience in GPU programming using CUDA or Triton kernels (or Palace/Mosaic kernels).
• Background in Compiler optimizations or related fields would also be beneficial.