Senior HPC Networking Architect: InfiniBand & Fortinet

Doist; United States; Remote; USD 140,000 - 180,000; Full time

Job summary

Doist is seeking a Senior HPC Networking Engineer to design, deploy, manage and troubleshoot high‑performance networks for HPC workloads, with InfiniBand fabrics and Fortinet security at the core.

You will diagnose InfiniBand and Ethernet issues, optimize fabric topology, and collaborate with compute, storage and platform teams to ensure reliability and scalability across hybrid clusters. You will also document architectures and participate in on‑call escalation.

Qualifications

• Extensive InfiniBand and HPC networking experience.
• Strong Ethernet networking and routing knowledge.
• Hands-on Fortinet security appliance experience.
• Experience designing, deploying, and maintaining HPC networks.

Responsibilities

• Design, deploy, and maintain high-performance HPC networks with InfiniBand.
• Troubleshoot InfiniBand and Ethernet issues to minimize downtime and optimize performance.
• Manage InfiniBand components: switches, HCAs, subnet managers, fabric configurations.
• Tune performance, monitor capacity, and plan scaling for HPC networks.
• Implement Fortinet security across networks using FortiGate/Manager/Analyzer.
• Collaborate with compute/storage/platform teams for HPC workloads.
• Document network architectures and operational procedures.
• Participate in on-call rotations and escalation for incidents.
• Lead or contribute to network upgrades, migrations, and new deployments.

Skills

InfiniBand

High-performance networking

Troubleshooting

HPC networking

Tools

Fortinet FortiGate

FortiManager

FortiAnalyzer

Job description

Doist is seeking a Senior HPC Networking Engineer to design, deploy, manage and troubleshoot high‑performance networks for HPC workloads, with InfiniBand fabrics and Fortinet security at the core.

You will diagnose InfiniBand and Ethernet issues, optimize fabric topology, and collaborate with compute, storage and platform teams to ensure reliability and scalability across hybrid clusters. You will also document architectures and participate in on‑call escalation.