AI Infrastructure Engineer

Main mission

Manages GPU clusters, distributed training, and computing costs.

5 key responsibilities

• Design and operate GPU training and inference clusters.
• Optimize computing usage: scheduling, parallelism, queues.
• Manage computing costs and trade-offs between cloud and on-premise.
• Ensure reliability of long jobs: checkpointing, recovery, monitoring.
• Equip ML teams to be autonomous on infrastructure.

Key skills

Kubernetes, Slurm, GPU (CUDA), high-performance networking, cloud, FinOps for computing.

What's expected

Fast, available computing at the right cost: infrastructure is the backbone of model warfare.

Career paths

AI Platform Architect, Head of Infrastructure, HPC Engineer.