Hybrid GPU Infra Engineer for ML Systems & Kubernetes

H Company; Paris; Híbrido; EUR 70 000 - 110 000; Tempo integral

Resumo da oferta

H Company is seeking an Infrastructure engineer to own GPU compute powering model training, evaluation and serving systems. You’ll run Kubernetes clusters for GPU training and inference and manage capacity with cloud providers to support our research, product and platform teams.

You’ll work with the Platform Infrastructure team, research and product groups, and our cloud providers to ensure reliability, scalability and efficient resource usage across regions.

Qualificações

• Experience building and running GPU workloads on Kubernetes.
• Ability to design a GPU cluster for training and online RL from scratch.
• Debugging OS, Linux, drivers, networking and storage at scale.
• Proficiency writing Python.

Responsabilidades

• Run Kubernetes clusters for GPU training and inference.
• Improve scheduling and placement of GPU work.
• Detect, drain and repair unhealthy GPU nodes.
• Manage capacity with cloud providers.
• Debug GPU, Linux, storage and networking failures.

Conhecimentos

Python

Kubernetes

OS debugging

Ferramentas

Slurm

Kubernetes scheduling tools (Kueue/Karpenter)

SkyPilot

AWS EKS/SageMaker HyperPod

NCCL/Open MPI

Terraform

Descrição da oferta de emprego

H Company is seeking an Infrastructure engineer to own GPU compute powering model training, evaluation and serving systems. You’ll run Kubernetes clusters for GPU training and inference and manage capacity with cloud providers to support our research, product and platform teams.

You’ll work with the Platform Infrastructure team, research and product groups, and our cloud providers to ensure reliability, scalability and efficient resource usage across regions.