This is a fully remote position, open to applicants in United States.

• Oversee the health and performance of InfiniBand and Ethernet networks: including switches, HCAs, transceivers, and links.

• Analyze and resolve fabric-related issues: such as connectivity problems, congestion, and performance regressions.

• Assist in fabric deployment in collaboration with data center operations and customer-facing teams.

• Conduct maintenance and upgrades on switches and control plane elements.

• Collaborate with cluster operations on cross-domain incidents where the boundaries between compute and network are unclear.

• Enhance tools and runbooks to ensure that the resolution of future incidents is quicker than before.

• Experience operating InfiniBand fabrics in a production environment: including subnet management, routing, partitioning, and monitoring.

• Proficient in debugging the entire stack: from cables and transceivers to switch firmware, HCAs, drivers, and NCCL.

• Successfully established new fabrics from cable installation through to validation.

• Created scripts to automate repetitive operational tasks (using bash, python, go, or similar languages).

• Preferred experience: Ethernet RoCE, Spectrum-X, or large-scale GPU cluster networking.

• Comprehensive health and wellness programs.

• Opportunities for professional development and training.

• Flexible work arrangements to promote work-life balance.

• Competitive salary and performance-based incentives.

Operations Engineer, HPC Networking

People also viewed