Research Engineer, Gemini Large Scale Pretraining, DeepMind

Overview

In this role you will work on long-term research-enabled projects with cross-functional teams. You will respond to time-sensitive issues in large-scale training and ongoing research, and improve technical workflows used across pretraining. You’ll apply engineering expertise to help researchers tackle technical challenges, balancing robust practices with a flexible, researcher-friendly codebase. You’ll operate with a broad view of the codebase, model architecture, and data pipelines to sustain scalable, impactful AI work. This position offers the chance to shape high-impact experiments and contribute to the wider research community.

Pay / Benefits

• 15% bonus target
• equity
• benefits
• opportunity to work across locations
• research collaboration opportunities

Responsibilities

• Advance long-term research projects with multiple team members
• Respond to time-sensitive issues in large-scale training runs and ongoing research
• Systematically improve technical workflows and tooling across pretraining
• Apply engineering expertise to help researchers navigate technical challenges
• Balance engineering practices with a flexible codebase that researchers can use and extend

Key requirements

• Bachelor’s in Computer Science or related field, or equivalent practical experience
• 2 years of software engineering experience with Python-based ML/scientific libraries (e.g., JAX) or OSS contributions
• 1 year of machine learning experience
• Experience with large-scale system design, distributed systems, or performance engineering (preferred)
• Strong understanding of engineering best practices, codebase design, and testing
• Familiarity with training modern large language models
• Ability to collaborate and communicate effectively across distributed, cross-skilled teams

• collaboration
• clear communication
• cross-functional teamwork
• Python
• JAX
• ML libraries

…