HPC Storage Engineer

TL;DR

HPC Storage Engineer (Parallel Filesystems): Designing, operating, and optimizing high-throughput storage platforms for demanding compute workloads with an accent on parallel filesystems like Lustre, GPFS, and BeeGFS. Focus on eliminating I/O bottlenecks, tuning metadata performance, and ensuring operational resilience for AI/ML and scientific workloads.

Company

hirify.global specializes in high-performance computing infrastructure and operational excellence.

What you will do

• Deploy and manage high-performance parallel filesystem services including Lustre, IBM Spectrum Scale (GPFS), and BeeGFS.
• Optimize I/O throughput, latency, and metadata performance through systematic tuning and data-path optimization.
• Design and evolve filesystem architectures, including MDS/MDT, OSS/OST, targets, pools, and tiering.
• Develop automation for repeatable builds, patching, and health checks using Ansible, Bash, and Python.
• Perform deep performance engineering using tools like IOR, mdtest, and fio to validate system changes.
• Collaborate with HPC, Linux, and networking teams to align storage behavior with real-world workload patterns.

Requirements

• 3–7+ years of hands-on experience operating HPC storage platforms and parallel filesystems in production.
• Proven expertise in installation and day-2 operations of Lustre, IBM Spectrum Scale (GPFS), or BeeGFS.
• Strong Linux systems administration skills (RHEL, Rocky, Ubuntu), including performance tuning and troubleshooting.
• Deep understanding of RDMA, InfiniBand, NVMe/SAS, and RAID fundamentals.
• Experience with storage observability, capacity planning, and systematic root-cause analysis.
• Fluent English (written and spoken) for cross-team collaboration and operational documentation.

Nice to have

• Experience with HPC schedulers like Slurm and their impact on storage throughput.
• Familiarity with object storage (S3) or HSM tiering concepts.
• Exposure to HPC container technologies such as Apptainer, Singularity, or Docker.