Software Engineer, Training Infrastructure

Job Description

Google DeepMind is seeking a Software Engineer to join their Training Infrastructure team. This role involves collaborating with a team of research engineers to enhance the tools and technology used for building large-scale training systems, particularly within the Gemini project. The engineer will contribute to critical research challenges, from technical roadmaps to infrastructure design and experiment analysis.

The role involves:

• Translating research requirements into technical roadmaps.
• Implementing and documenting research infrastructure.
• Upskilling and contributing to the research agenda.
• Supporting the growth of junior team members.
• Contributing to team culture and promoting sustainability.

Requirements include:

• Bachelor's degree or equivalent practical experience.
• 8 years of experience in software development with data structures/algorithms.
• 5 years of experience building, testing, and supporting software in research.
• Experience building large-scale infrastructure for research in Deep Learning.
• Understanding of accelerators (e.g., Jax & XLA stack), performance profiling, and optimization.
• Experience with distributed systems, resilience, and performance.
• Strong communication skills and comfort working in a distributed team.

Google DeepMind offers:

• Bonus
• Equity
• Benefits