Network Engineer (AI Infrastructure)

Location: San Francisco, CA / Remote within the US; travel up to 25%.

Company

hirify.global is a market and infrastructure platform for buying, selling, operating, and financing compute, serving model builders, research labs, inference providers, and compute infrastructure providers.

What you will do

• Design and bring up compute fabrics for large-scale GPU clusters, including InfiniBand, Spectrum-X Ethernet, rail-optimized topologies, and NVLink domain integration.
• Design and capacity-plan storage, tenant-access, internet, and cloud-interconnect networks.
• Create cabling plans, optics BOMs, and labeling standards for integrators.
• Perform fabric bring-up, NCCL and fabric performance testing, and acceptance sign-off.
• Serve as the escalation point for deployment and Day 2 fabric issues.
• Develop configuration templates and tooling that establish a known-good baseline for each build.

Requirements

• Hands-on design and bring-up of InfiniBand or RoCE fabrics for clusters with 1,000+ GPUs.
• Experience with NVIDIA UFM and switch operating systems such as Cumulus or SONiC.
• Practical knowledge of optics, transceivers, and high-density cabling.
• Experience using Python and Ansible or similar tools for configuration and validation.
• Ability to work with the Cluster Architect on design and deployment engineers on site through validation.
• Ability to travel up to 25%.

Nice to have

• Experience with GB200 or GB300 NVL72 rack-scale systems.
• Data center interconnect and cloud interconnect design experience.
• Experience working directly with customer ML infrastructure teams.

Culture & Benefits

• Ownership of the network design standard for GPU clusters.
• Work on current and next-generation fabrics across multiple sites and customers.
• High-trust environment with a high talent density.
• Healthcare, dental, and vision coverage for employees and dependents.
• 401(k) and unlimited PTO.