HPC Systems Engineer: GPU Clusters & Automation

Susquehanna International; Bala Cynwyd (PA); On-site; USD 120,000 - 180,000; Full time

Job summary

Susquehanna International is seeking an HPC Engineer to build and optimize our trading and research systems. You will manage thousands of GPUs, huge compute clusters, and petabytes of storage, collaborating with researchers and traders to deliver high-performance infrastructure.

You will automate monitoring, tune kernels and I/O, and integrate solutions across storage, networking, and scheduling. The role emphasizes scalability, reliability, and deep systems troubleshooting.

Qualifications

• Bachelor’s degree or equivalent experience in a relevant field.
• Experience building and operating Linux-based HPC platforms.
• Experience tuning kernel and I/O subsystems and troubleshooting performance (sysctl, perf, strace, tcpdump, Wireshark).
• Strong automation skills in Python, Bash, or Ansible.
• Experience administering parallel filesystems Lustre/GPFS/VAST.
• Experience administering and tuning Slurm workload scheduler.
• Familiarity with InfiniBand/RoCE and GPU cluster environments.

Responsibilities

• Automate and evolve tools to monitor and maintain performance.
• Collaborate with Strategy Developers, Quant Researchers, and IT teams to build scalable solutions.
• Optimize OS and batch workflows; perform root-cause analysis of issues.
• Oversee HPC environment: scheduler, filesystems, GPUs, interconnects.
• Design and optimize high-throughput storage solutions (Lustre, VAST, GPFS).
• Plan capacity and forecast resource needs using analytics.
• Troubleshoot and tune systems to ensure timely recovery of workflows.

Skills

Linux HPC platforms

Python automation

Bash scripting

Ansible/config mgmt

GPU/InfiniBand interconnects

Education

Bachelor’s degree in Engineering/CS/Information Systems

Tools

Lustre

GPFS

VAST

Slurm

Job description

Susquehanna International is seeking an HPC Engineer to build and optimize our trading and research systems. You will manage thousands of GPUs, huge compute clusters, and petabytes of storage, collaborating with researchers and traders to deliver high-performance infrastructure.

You will automate monitoring, tune kernels and I/O, and integrate solutions across storage, networking, and scheduling. The role emphasizes scalability, reliability, and deep systems troubleshooting.