HPC & AI Platform Engineer — GPU Clusters & Automation

Era4; England; On-site; GBP 70,000 - 110,000; Full time

Job summary

Era4 develops and operates AI infrastructure across the UK, turning brownfield sites into modern data-centre facilities for healthcare, research, finance, enterprise and public sector organisations.

The Platform Engineer role shapes Era4's new Platform team, is customer-facing, and involves troubleshooting and cross-vendor collaboration to ensure seamless AI platform operations at scale.

Qualifications

• Experience supporting AI/HPC infrastructure and platforms.
• Linux system administration with RHEL/CentOS and Ubuntu.
• Proficiency in NVIDIA CUDA toolkits, Kubernetes, and Ansible.
• Familiarity with high-performance networking and security hardening.
• Ability to troubleshoot, automate, and collaborate with vendor teams.

Responsibilities

• Design, deploy, and manage large-scale HPC and GPU clusters.
• Implement HPC scheduling and resource management (Slurm).
• Architect InfiniBand/Ethernet topologies and optimise networks.
• Ensure high availability and proactive risk mitigation.
• Automate provisioning, monitoring, and workflows across multi-vendor stacks.
• Lead real-time performance monitoring and fault diagnosis with vendor support.
• Manage security, RBAC, and data protection; interface with customers.

Skills

AI/HPC platform experience

System administration

Linux admin

NVIDIA CUDA toolkits

Kubernetes & containers

Automation with Ansible

HPC networking

Cloud APIs

Monitoring Grafana

Security RBAC

Vendor collaboration

Tools

Ansible

NVIDIA CUDA toolkit

Kubernetes

Grafana

Kibana

Splunk

Job description

Era4 develops and operates AI infrastructure across the UK, turning brownfield sites into modern data-centre facilities for healthcare, research, finance, enterprise and public sector organisations.

The Platform Engineer role shapes Era4's new Platform team, is customer-facing, and involves troubleshooting and cross-vendor collaboration to ensure seamless AI platform operations at scale.