Network Architecture Engineer

About Mavria

Mavria’s mission is to build the industrial capacity America needs to lead in AI. We believe that sustaining America’s AI leadership requires an unprecedented expansion of power, manufacturing, supply chains and physical infrastructure.

Our founders are serial entrepreneurs who met at the U.S. Department of Energy, working on some of the country’s hardest energy and infrastructure projects. There, they saw a widening gap: AI labs and cloud providers were racing to scale compute, while the infrastructure underneath them was still built one custom project at a time, on timelines measured in years.

We founded Mavria to close that gap. We are building to industrialize intelligence: turning the physical infrastructure AI runs on into products we manufacture, deploy and improve with every build.

We’re starting with data centers. Our first product, Terrain, is a factory-built data center platform. Terrain 01 is being built in Houston now, and we’ve raised $130M as of October 2026 to accelerate our roadmap.

Terrain is our first product, not the limit of our ambition. We intend to eliminate the physical bottlenecks that constrain American AI leadership and build one of the foundational infrastructure companies of the AI era.

How we build

At Mavria, three principles guide how we work:

• Ambition: We want to build a generational company. To do that, we build capabilities, not just products. Terrain is our first product, not our last – and we seek to eliminate strategic bottlenecks across the entire AI supply chain.

• Urgency: AI moves in weeks, while infrastructure tends to move in years. We match the frontier’s pace by making decisions quickly, building in parallel and securing critical supply ahead of demand. We surface problems early and never compromise on safety or engineering rigor.

• Ownership: When we say we’ll do something, we do it. Keeping our word is how we stand apart, and taking radical ownership over our scope and our domain is how we build trust – internally and externally – while moving fast.

Role scope

As our first dedicated network architecture hire, you’ll design, deploy and run the data center network behind every Terrain site: how servers, switches and cabling sit in each rack, the leaf-spine fabric that connects racks, rows and halls, and the long-haul circuits that connect each site to the outside world. The standards you set will become the template for every site we build. You will:

• Design the fabric: Design and evolve Ethernet leaf-spine fabrics for AI and GPU workloads, with redundancy so no single failure takes a site down.

• Design from the rack up: Lay out rack-level networks, including top-of-rack or end-of-row, port density, patching, fiber or copper, and how network and power sit together in the cabinet, plus the structured cabling and fiber plant across rows and halls.

• Connect the sites: Own long-haul and data center interconnect circuits, carrier handoffs and storage network integration.

• Deploy and run it: Deploy, configure and manage network infrastructure across our campus and sites, keep install packages, as-builts and the network asset database current, and plan upgrades, migrations and decommissioning.

• Set the standard: Define standards for configuration, routing, congestion management, firmware lifecycle and safe change management, and test new hardware and software platforms.

• Own the incidents: Act as the escalation point for network incidents, drive permanent fixes, and keep improving uptime, latency and capacity.

• Build it into the product: Work with our manufacturing, MEP and integration teams so the network design fits each standardized unit and site, including power and cooling.

What we're looking for

For this role, we’re looking for a hands-on data center network engineer with the following qualities:

• A data center builder: You have 6+ years designing and running large-scale data center networks, and you’ve physically built and logically designed leaf-spine fabrics in production, not only office or enterprise networks.

• Rack-literate: You know what an IT rack looks like, how it’s wired, and how to connect hundreds of them into a resilient fabric.

• Deep on protocols: You work with BGP, IS-IS, OSPF, MPLS, ECMP, EVPN-VXLAN and IPv4/IPv6 across multi-vendor platforms such as Arista, NVIDIA Spectrum, Juniper or Cisco.

• Optics-fluent: You’ve designed fiber plants and worked with 100G-and-above optics, CWDM/DWDM and coherent optics.

• A clear explainer: You can explain complex design choices in plain language to people who aren’t network experts.

• Ready for the field: You’re comfortable with a 24x7 on-call rotation and 10–30% travel to sites, data centers and vendors.

AI or GPU cluster networking experience is especially relevant, as are field work in data centers, colocation facilities or internet exchanges, an understanding of data center power and cooling, and automation in Python or Bash. We value ability over years or titles.

To apply: Share your background and an example of the largest leaf-spine fabric you designed or built.

Benefits

• Competitive total compensation package

• Medical, dental, and vision insurance is 100% paid for the employee

• Retirement Plan with a 3.5% company match

• Flexible time off

Mavria is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, sexual orientation, gender identity, age, disability, protected veteran status, or any other characteristic protected by law. Mavria will consider qualified applicants with arrest and conviction records in accordance with applicable law.