Software Engineer, Model Inference, DeepMind

Software Engineer, Model Inference, DeepMind Overview

As a Software Engineer in Model Inference at Google DeepMind, you will play a crucial role in advancing AI technology by collaborating with a talented team of researchers and engineers. Your work will focus on optimizing and deploying large language models (LLMs) like Gemini onto Google’s production infrastructure, impacting a wide range of applications.

Responsibilities

• Collaborate with research teams to understand and implement next-generation modeling approaches with production considerations.
• Work alongside infrastructure teams to create serving infrastructure optimized for efficiency and performance, addressing speed, scale, and quality bottlenecks.
• Identify automation opportunities to streamline tasks, reduce redundancies, and enhance the speed of model releases.
• Develop a comprehensive understanding of serving frameworks, pre-processing pipelines, caching mechanisms, and related technologies.
• Utilize roofline analysis, hardware-level profiling, and systems analysis to detect and resolve performance bottlenecks across ML frameworks, compilers, custom kernels, and serving infrastructure on hardware accelerators.

Requirements

• A Bachelor’s degree or equivalent practical experience.
• At least 8 years of experience in software development.
• Minimum of 2 years of experience in deploying and maintaining machine learning models in a live production environment.
• Experience with profiling, configuring, or executing machine learning workloads on hardware accelerators such as GPUs or TPUs.
• Proven experience in designing, building, or optimizing model serving infrastructure or inference backends.
• Familiarity with developing serving infrastructure and programming hardware accelerators using ML frameworks or low-level programming models.
• Experience in profiling software to identify performance bottlenecks and optimizing distributed ML systems.

Benefits

• Competitive salary ranging from $207,000 to $300,000 (USD) plus a 20% bonus target and equity.
• Comprehensive benefits package including health, wellness, and retirement plans.
• Opportunities for professional development and career advancement within a pioneering AI lab.

This role is hybrid, with options to work from London, UK, or Mountain View, CA, USA. Please note that this position is not eligible for visa sponsorship.

Degree Requirement: Not Specified

Visa Sponsorship May be