About the job
We engineer the future of Artificial Intelligence (AI) serving infrastructure at the intersection of Large Language Models (LLMs) and high-performance computing. Our team drives foundational gains in efficiency, latency, and throughput to scale Google’s most advanced models globally. In this role, you will build tools to maximize Large Language Model (LLM) performance on cutting-edge Graphics Processing Unit (GPU) platforms. You will develop next-generation disaggregated serving architectures and enable the seamless deployment of Gemini across Google’s products and Cloud infrastructure.
Responsibilities
Build infrastructure and tooling for deep profiling, benchmarking, and analysis of large language models (LLMs) on graphics processing unit (GPU) accelerators.
Identify and resolve bottlenecks across compute, memory, and networking to maximize hardware efficiency.
Prototype serving techniques, including disaggregated serving, speculative decoding, and optimized key-value (KV) cache management.
Design and implement enhancements to the serving stack to improve latency, throughput, and resource utilization.
Partner with research, engineering, and site reliability engineering (SRE) teams to deploy models into production.
Qualifications
Minimum
Bachelor's degree or equivalent practical experience.
8 years of experience in software development.
5 years of experience testing and launching software products.
5 years of experience with performance, large scale systems data analysis, visualization tools, or debugging.
3 years of experience with software design and architecture.
3 years of experience in low-level systems optimization, including performance tuning for GPU or TPU accelerators or high-performance distributed AI serving infrastructure.
Preferred
Master’s degree or PhD in engineering, computer science, or a related technical field.
8 years of experience with data structures and algorithms.
3 years of experience in a technical leadership role leading project teams and setting technical direction.
3 years of experience working in a matrixed organization involving cross-functional or cross-business projects.
3 years of experience leading the architecture and multi-quarter technical roadmap for AI inference or training systems, focusing on hardware-software co-design.
3 years of experience implementing optimization techniques, such as quantization, speculative decoding, or memory management to reduce total cost of ownership.