24/7 Linux HPC Operations Engineer (RDMA & Storage)

Jump Trading; Sydney; On-site; AUD 120,000 - 180,000; Full time

Job summary

Jump Trading Group is seeking an HPC Operations Engineer to support 24/7 Linux HPC compute, storage, and interconnects in Sydney and Mumbai. You will respond to alerts, resolve issues for researchers, and participate in large maintenance windows.

You will write code to diagnose and automate tasks and work across teams with diverse technologies. Ideal candidates have 2+ years of Linux experience, programming/scripting skills, and strong communication.

Qualifications

• At least 2+ years of professional experience with Linux systems.
• Experience with HPC concepts is a plus (not required).
• Proficient in at least one programming/scripting language (Go, Python, C).
• Strong ability to perform root-cause analysis and communicate findings clearly.

Responsibilities

• Provide 24/7 operational support for Linux HPC compute, storage, and interconnects.
• Resolve issues from Jump's research community and manage problem lifecycles.
• Respond to alerts and participate in on-call rotations.
• Contribute to maintenance windows, evenings, and weekend work as needed.
• Develop tooling to diagnose, resolve, and automate common tasks.
• Collaborate across teams to build testing infrastructures for new and existing codebases.
• Coordinate with vendors during domestic and international meetings.

Skills

Linux expertise

Programming/scripting

Root-cause analysis

Communication

Project management

Tools

Lustre

GPFS

Slurm

Grid Engine

Job description

Jump Trading Group is seeking an HPC Operations Engineer to support 24/7 Linux HPC compute, storage, and interconnects in Sydney and Mumbai. You will respond to alerts, resolve issues for researchers, and participate in large maintenance windows.

You will write code to diagnose and automate tasks and work across teams with diverse technologies. Ideal candidates have 2+ years of Linux experience, programming/scripting skills, and strong communication.