GPU Infrastructure Engineer (AI)

Location: Hybrid in Paris or London, with three days per week in the office and occasional travel between the two

Company

hirify.global develops AI systems that depend on large-scale GPU compute for model training, evaluation, and serving.

What you will do

• Run Kubernetes clusters supporting GPU training and inference.
• Improve GPU scheduling and workload placement across training and serving systems.
• Detect, drain, and repair unhealthy GPU nodes.
• Manage compute capacity with cloud providers, including capacity expansion and new regions.
• Debug failures across GPUs, Linux, storage, networking, and related infrastructure.
• Own a portion of the compute layer and help operate a globally distributed GPU system.

Requirements

• Experience building and running GPU workloads on Kubernetes.
• Ability to design a GPU cluster for training and online reinforcement learning from scratch.
• Strong debugging skills at the operating-system layer and below, including Linux, GPU drivers, networking, and storage.
• Strong Python programming skills.
• Ability to work hybrid from the Paris or London office three days per week and travel occasionally between the offices.

Nice to have

• Experience as an SRE or infrastructure engineer.
• Experience with Slurm, Kueue, Karpenter, SkyPilot, AWS EKS, or SageMaker HyperPod.
• Experience with EFA, InfiniBand, NCCL, Open MPI, or Terraform.

Culture & Benefits

• Work with platform infrastructure, research, product, cloud provider, and vendor stakeholders.
• Responsibilities expand from incident handling and infrastructure changes to ownership of a complete compute-layer area.
• Competitive compensation package.

Hiring process

• 30-minute conversation with the Talent team.
• 60-minute coding interview and 60-minute system design interview based on real-world role problems.
• 30-minute closing conversation with the VP of Engineering.