Senior HPC & Cloud Engineer - AI/ML Compute Infrastructure

Senior HPC & Cloud Engineer - AI/ML Compute Infrastructure
\n Oxford | Hybrid (3 days in office) | Competitive, DOE

Sentinel is recruiting for several senior/staff-level engineers to design, build and operate a hybrid GPU compute environment, combining on-prem HPC clusters with public cloud infrastructure for large-scale AI research workloads.

Responsibilities:

\n
• Build and operate high-performance GPU training/inference clusters, including scheduling, isolation and automated life cycle management
\n
• Design high-throughput data paths across compute and storage, including parallel filesystems (eg Lustre)
\n
• Benchmark and resolve performance bottlenecks across compute, network and orchestration layers
\n
• Implement observability, resilience and security controls for a compliance-conscious research environment
\n
• Work with research and applied ML teams to forecast GPU/storage capacity and streamline experimentation pipelines
\n

Requirements:

\n
• Experience with HPC/GPU clusters, including a strong understanding of GPU architecture, high-speed networking and distributed training performance
\n
• Cloud platform experience (Azure, GCP, AWS or other)
\n
• Experience with containerisation/Kubernetes
\n
• Working knowledge of IaC and CI/CD (eg Terraform, Argo CD)
\n