HPC Platform Management Engineer

Qube Research & Technologies is seeking a Technologist to join their Workload Scheduling (WLS) team. This role focuses on developing, deploying, and maintaining HPC platforms using Yellow Dog and Ray schedulers. The ideal candidate will be instrumental in integrating HPC solutions, improving performance, and contributing to infrastructure automation.

Requirements

• Experience of engineering and supporting at least one HPC scheduler
• Good understanding of loosely and tightly coupled HPC workloads
• Experience of developing and supporting large-scale systems (5000+ nodes) and high levels of concurrency (100k+ tasks)
• Experience of monitoring and visualisation of large-scale systems
• Performance tuning of compute, network and storage components
• Experience of using AWS IAM and Okta for user authorisation
• Solid understanding of Linux administration skills
• Good understanding of various storage solutions
• Effective communication skills
• Proactive and collaborative approach