Scale & Infrastructure Engineer for Large-Scale AI

AI Breaking Wire; Mountain View, Northern (CA, KY); Hybrid; USD 165,000 - 230,000; Full time

Benefits offered by this job

Stock options

Healthcare coverage

PTO

Retirement match

Job summary

Google DeepMind is seeking a Research Engineer for the Infrastructure and Scale team. You will build foundational systems that train state‑of‑the‑art AI models and scale computation across large TPU clusters.

Responsibilities include designing high‑throughput distributed training frameworks, profiling performance bottlenecks on GPUs/TPUs, and collaborating with researchers to co‑design new features and system capabilities.

Qualifications

• Strong systems background with distributed computing experience.
• Proficiency in C++ and Python is required.
• Experience with MPI, JAX or TensorFlow is preferred.

Responsibilities

• Architect and optimize high-throughput distributed training frameworks for large models.
• Profile and debug performance bottlenecks across hardware accelerators and software stacks.
• Collaborate with researchers to co-design new algorithmic features and system capabilities.

Skills

Strong systems background

C++

Python

Distributed systems

Education

Bachelor's degree or higher in CS/EE or related

Tools

MPI

JAX

TensorFlow

Job description

Google DeepMind is seeking a Research Engineer for the Infrastructure and Scale team. You will build foundational systems that train state‑of‑the‑art AI models and scale computation across large TPU clusters.

Responsibilities include designing high‑throughput distributed training frameworks, profiling performance bottlenecks on GPUs/TPUs, and collaborating with researchers to co‑design new features and system capabilities.