HPC & AI Platform Engineer – GPU Clusters

Era4; United Kingdom; Hybrid; GBP 95,000 - 130,000; Full time

Job summary

Era4 is seeking a Platform Engineer (HPC & AI) in the United Kingdom to shape a new Platform team. The role is customer facing, involves troubleshooting, and collaboration with vendor engineering teams to ensure seamless AI platform operations.

You will design, deploy, and manage large-scale HPC and GPU-accelerated clusters (NVIDIA). You will implement Slurm-based scheduling, optimize InfiniBand/Ethernet networks, and lead automation, monitoring, and incident response across multi-vendor stacks

Qualifications

• Experience supporting AI/HPC infrastructure and platforms.
• System administration experience with Linux (RHEL/CentOS, Ubuntu).
• Proficiency with Ansible, Nvidia CUDA toolkits, Kubernetes; automation and security focus.

Responsibilities

• Design, deploy, and manage large-scale HPC and GPU-accelerated clusters (NVIDIA).
• Implement and administer HPC scheduling with GPU partitioning (e.g., Slurm).
• Architect and optimise InfiniBand and Ethernet network topologies.
• Ensure high availability and resilience with failover and maintenance planning.
• Automate provisioning, configuration, monitoring, and workflows across multi-vendor stacks.
• Monitor real-time performance and lead troubleshooting with vendor support.
• Incident response: node/network/driver issues and escalation with vendors.
• Security and access control: manage permissions, RBAC, data protection.

Skills

HPC/AI infra

Linux admin

Ansible

NVIDIA CUDA

Kubernetes

GPU scheduling

SLURM

High perf networking

RBAC security

Tools

NVIDIA CUDA toolkit

Kubernetes

Ansible

Job description

Era4 is seeking a Platform Engineer (HPC & AI) in the United Kingdom to shape a new Platform team. The role is customer facing, involves troubleshooting, and collaboration with vendor engineering teams to ensure seamless AI platform operations.

You will design, deploy, and manage large-scale HPC and GPU-accelerated clusters (NVIDIA). You will implement Slurm-based scheduling, optimize InfiniBand/Ethernet networks, and lead automation, monitoring, and incident response across multi-vendor stacks