Senior Solutions Architect – AI Infrastructure Networking

• Own the network solutioning of k0rdent AI, Mirantis's platform for building and operating GPU clouds and AI factories
• Design large GPU-cluster interconnects, tenant isolation, and workload connectivity from containers, virtual machines and bare-metal nodes
• Design and publish network reference architectures and solution designs from single-rack to multi-thousand-GPU clusters
• Define InfiniBand and RoCEv2 Ethernet compute fabrics, including rail-optimised and fat-tree/Clos designs, oversubscription and failure domains
• Define front-end, storage and management networks and connectivity to customer data centers, public clouds and hybrid environments
• Specify multi-tenant isolation using InfiniBand PKeys, VRFs, EVPN-VXLAN and Kubernetes network policy
• Document scale limits and trade-offs involving cost, performance, operability and vendor lock-in
• Define Linux host networking for NICs, DPUs and SuperNICs, including PCI passthrough, SR-IOV, IOMMU, NUMA and GPU-NIC affinity
• Design Kubernetes networking for AI workloads and KubeVirt virtual machines
• Research emerging AI-networking technologies and standards, prototype alternatives and benchmark them against current practice
• Publish internal research notes, design proposals, papers, blog posts and talks
• Build and run customer and partner proofs of concept on Mirantis lab hardware and customer sites
• Write automation and tooling using Python, Go, Bash, Ansible, Helm, Kubernetes manifests and Terraform
• Validate and benchmark fabrics and host configurations using NCCL tests, perftest and ib_write_bw
• Act as the network subject matter expert in customer discovery, design reviews and architecture workshops
• Collaborate with hardware/networking partners, Solutions Architects, Partner Management, Engineering, Product and customers across time zones
• Present at industry events, webinars and partner summits; run hands-on technical workshops
• Produce reference architectures, solution briefs, blog posts and internal enablement materials
• Help shape the k0rdent AI product roadmap

• Bachelor's degree in Computer Science, Electrical Engineering, Telecommunications or a related field, or equivalent practical experience
• 8+ years in network engineering or network architecture
• At least 3 years in data center, HPC or cloud infrastructure networking
• Customer-facing experience as a solutions architect, pre-sales engineer, consultant or technical lead
• Data center network design experience including spine-leaf and Clos topologies, rail-optimised GPU fabrics, oversubscription, ECMP, and scaling to thousands of nodes
• InfiniBand expertise including subnet management, partitioning, adaptive routing, and NCCL/RDMA traffic behavior
• RoCEv2 expertise including PFC, ECN, DCQCN, QoS, MTU and buffer tuning
• Knowledge of BGP, EVPN-VXLAN and VRFs for multi-tenancy
• Hybrid and multi-cloud connectivity experience, including interconnects, VPN, cloud networking constructs, IP address and DNS planning
• Linux networking knowledge including NIC drivers, PCI passthrough, SR-IOV, IOMMU, NUMA affinity, iproute2, ethtool and devlink
• Kubernetes networking knowledge including CNI plugins, Multus, secondary networks, SR-IOV and RDMA device plugins, network policy and service exposure
• KubeVirt and VM networking, passthrough and SR-IOV configuration experience
• Programming or scripting in at least one language; Python or Go preferred
• Comfortable using Git, CI and infrastructure-as-code
• Excellent written and spoken English
• Comfortable presenting to large audiences and running hands-on workshops
• Able to adapt technical messaging to network engineers and customer executives
• Experience working in an international, distributed company across time zones and work cultures
• Willingness to work some meetings outside standard local hours
• Nice-to-have experience with NVIDIA networking, GPU cloud/neocloud/HPC operators, bare-metal provisioning, network automation, SDN controllers, switch operating systems, global load balancing, software-defined storage networking, Mirantis products, open-source contributions, industry standards, published research, patents, white papers or additional languages

• Remote work arrangement
• Travel of up to 25% for customer engagements, partner meetings, lab work and industry events
• Collaboration with a world-class, distributed team
• Opportunities to work directly with leading GPU cloud operators, NeoClouds, sovereign clouds, and AI-first enterprises
• Opportunity to shape the product narrative and influence go-to-market success