Senior Network Architect, AI Cloud & GPU Fabrics (Remote)

Mirantis, Inc.; Oregon; Hybrid; USD 170,000 - 230,000; Full time

Benefits offered by this job

Remote work in US

Travel up to 25% for customer/partner/

Professional development opportunities

Job summary

Mirantis is seeking a Senior Solutions Architect to own the network solutioning of k0rdent AI, from metal to model. You will design interconnects for large GPU clusters, tenant isolation, and how workloads flow across containers, VMs and bare metal nodes.

You will publish reference architectures, build proofs of concept in the Mirantis lab, and translate complex networking choices into customer-ready designs and foster collaboration across global teams.

Qualifications

• Bachelor's degree or equivalent practical experience.
• 8+ years in network engineering or data center/cloud infrastructure networking.
• Customer-facing experience as a solutions architect, pre-sales engineer or technical lead.

Responsibilities

• Design and publish network reference architectures for k0rdent AI, from a single rack to multi-thousand-GPU clusters.
• Define compute fabrics on InfiniBand and RoCEv2 Ethernet with rail-optimised and fat-tree/Clos designs.
• Specify multi-tenant isolation across the stack with InfiniBand partitions, VRFs and EVPN‑VXLAN.
• Document scale limits and trade-offs with defensible reasoning for customers.
• Build proofs of concept and write automation/tooling (Python/Go/Bash/Ansible) to reproduce references.
• Collaborate with partners and customers, present findings, and publish internal/external content.

Skills

Data center design

InfiniBand

RoCEv2 Ethernet

Kubernetes networking

Python/Go scripting

Education

Bachelor's degree in CS/EE/telecom or equivalent

Tools

Cilium

Calico

OVN-Kubernetes

Multus

SR-IOV

RDMA device plugins

Job description

Mirantis is seeking a Senior Solutions Architect to own the network solutioning of k0rdent AI, from metal to model. You will design interconnects for large GPU clusters, tenant isolation, and how workloads flow across containers, VMs and bare metal nodes.

You will publish reference architectures, build proofs of concept in the Mirantis lab, and translate complex networking choices into customer-ready designs and foster collaboration across global teams.