ML Inference & Serving Engineer

Google Inc.; Greater London; Hybrid; GBP 153,000 - 222,000; Full time

Job summary

Google DeepMind in London, UK and Mountain View, CA is seeking a Software Engineer for model inference. You will work with researchers to optimize and deploy large language models (LLMs) on Google's production infrastructure, building scalable serving backends and performance-driven tests.

The role welcomes both IC and TL paths and offers opportunities across multiple teams. A strong software engineering foundation and experience with ML workloads on accelerators is essential.

Qualifications

• Bachelor's degree or equivalent practical experience.
• 8 years of experience in software development.
• 2 years of experience in deploying and maintaining machine learning models in a live production environment.
• Experience in profiling, configuring, or executing ML workloads directly on hardware accelerators (e.g., GPU or TPU).
• Experience designing, building, or optimizing model serving infrastructure or inference backends.

Responsibilities

• Create systems for agent testing in 2D and 3D games and develop test problems within physics simulators.
• Develop graphical visualizations of results and build competitive agent leaderboards.
• Test new algorithms on robots and collaborate with ML/ neuroscience teams.
• Optimize and deploy large language models (LLMs) like Gemini on production infrastructure.

Skills

Software development

ML deployment

Profiling on GPUs/TPUs

Model serving infrastructure

Systems optimization

Education

Bachelor's degree or equivalent practical experience

Tools

JAX

PyTorch

CUDA

OpenCL

Pallas

XLA

Job description

Google DeepMind in London, UK and Mountain View, CA is seeking a Software Engineer for model inference. You will work with researchers to optimize and deploy large language models (LLMs) on Google's production infrastructure, building scalable serving backends and performance-driven tests.

The role welcomes both IC and TL paths and offers opportunities across multiple teams. A strong software engineering foundation and experience with ML workloads on accelerators is essential.