Tech Lead, AI Inference Performance

Google; Greater London; Presencial; GBP 120 000 - 180 000; Tempo integral

Resumo da oferta

Google is seeking a senior leader to set the technical roadmap for a team of performance engineers, focusing on optimizing ML frameworks, kernels, and serving infrastructure. You will collaborate with researchers to balance latency, memory, and cost, guiding impactful optimizations across large-scale systems.

The role requires substantial experience in software leadership, performance analysis, and deep knowledge of Transformer/MoE architectures, with a push to deliver high-impact improvements

Qualificações

• Bachelor’s degree or equivalent practical experience.
• 8 years of experience in software development.
• 5 years in people management/leadership roles.
• 5 years in technical leadership overseeing projects.

Responsabilidades

• Set the technical roadmap for a team of performance engineers and drive prioritization of optimization opportunities.
• Leverage roofline analysis, hardware-level profiling, and systems analysis to identify bottlenecks across ML frameworks and serving infrastructure.
• Apply first-principles understanding of Transformer and Mixture-of-Experts components to optimize execution efficiency and memory footprint.
• Collaborate with research teams to evaluate inference implications and model how architecture affects latency and serving costs.
• Guide development of custom kernels and serving optimizations; conduct deep profiling of accelerator utilization and memory bandwidth.

Conhecimentos

Performance engineering

Team leadership

ML frameworks

Profiling tools

Transformer

Mixture of Experts

Formação académica

Bachelor's degree

Master's or PhD in Engineering/CS

Ferramentas

Roofline analysis

Hardware profiling tools

ML frameworks

Custom kernels

Descrição da oferta de emprego

Google is seeking a senior leader to set the technical roadmap for a team of performance engineers, focusing on optimizing ML frameworks, kernels, and serving infrastructure. You will collaborate with researchers to balance latency, memory, and cost, guiding impactful optimizations across large-scale systems.

The role requires substantial experience in software leadership, performance analysis, and deep knowledge of Transformer/MoE architectures, with a push to deliver high-impact improvements