Senior Solutions Architect – AI Infrastructure Networking

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

📋 Description

• Take ownership of network solutions for k0rdent AI, Mirantis's platform designed for constructing and managing GPU clouds and AI factories.

• Create designs for large GPU cluster interconnects, ensuring tenant isolation and workload connectivity across containers, virtual machines, and bare-metal nodes.

• Develop and share network reference architectures and solution designs that range from single-rack setups to multi-thousand-GPU clusters.

• Specify InfiniBand and RoCEv2 Ethernet compute fabrics, along with front-end, storage, and management networks.

• Outline multi-tenant isolation using InfiniBand partitions, VRFs, EVPN-VXLAN, and Kubernetes network policies.

• Document scalability limits and the trade-offs related to cost, performance, operability, and vendor lock-in.

• Define Linux host networking for various components including NICs, DPUs, SuperNICs, PCI passthrough, SR-IOV, IOMMU, NUMA, and GPU-NIC affinity.

• Develop Kubernetes networking solutions for AI workloads utilizing CNIs, Multus, SR-IOV, RDMA device plugins, NVIDIA Network Operator, and DRA.

• Construct KubeVirt networking for GPU and RDMA traffic.

• Investigate emerging AI networking technologies and standards, prototype alternative designs, and conduct benchmarking.

• Publish research notes, design proposals, papers, blog posts, and presentations.

• Create and manage proofs of concept with customers, partners, and on Mirantis hardware or customer sites.

• Write automation and tooling scripts using Python, Go, Bash, Ansible, Helm, Kubernetes manifests, and Terraform.

• Validate and benchmark fabrics and host configurations utilizing NCCL tests, perftest, and ib_write_bw.

• Serve as a network subject matter expert during customer discovery, design reviews, and architecture workshops.

• Collaborate with hardware and networking partners, contributing research to Product, Engineering, and the k0rdent AI roadmap.

• Present at industry events, webinars, and partner summits, as well as conduct technical workshops.

• Write reference architectures, solution briefs, blog posts, and internal enablement materials.

⛳️ Requirements

• Bachelor’s degree in Computer Science, Electrical Engineering, Telecommunications, or a related field, or equivalent practical experience.

• Over 8 years of experience in network engineering or network architecture.

• A minimum of 3 years in networking for data center, HPC, or cloud infrastructure.

• Customer-facing experience in roles such as solutions architect, pre-sales engineer, consultant, or technical lead.

• Proficiency in data center network design, including spine-leaf and Clos topologies, rail-optimized GPU fabrics, oversubscription, and ECMP.

• In-depth knowledge of InfiniBand, including subnet management, partitioning, adaptive routing, and NCCL/RDMA traffic.

• Expertise in RoCEv2 Ethernet, encompassing PFC, ECN, DCQCN, QoS, MTU, and buffer tuning.

• Familiarity with BGP, EVPN-VXLAN, VRFs, hybrid and multi-cloud connectivity, VPNs, IP address, and DNS planning.

• Strong Linux networking knowledge, covering NICs, PCI passthrough, SR-IOV, IOMMU, NUMA affinity, iproute2, ethtool, and devlink.

• Knowledge of Kubernetes networking, including CNI plugins, Multus, SR-IOV and RDMA device plugins, network policy, and service exposure.

• Experience with KubeVirt VM networking, passthrough, and SR-IOV.

• Proficient in programming or scripting in at least one language; Python or Go is preferred.

• Comfortable utilizing Git, CI, and infrastructure-as-code methodologies.

• Excellent proficiency in written and spoken English.

• Comfortable presenting to large audiences and conducting hands-on workshops.

• Experience working in an international, distributed company across various time zones and cultures.

• Willingness to attend some meetings outside of standard local hours.

• Nice-to-have experience includes NVIDIA networking, GPU cloud/HPC deployments, bare-metal provisioning, network automation, SDN, switch operating systems, GSLB, Gateway API, NVMe-oF, GPUDirect Storage, Mirantis products, open-source contributions, industry standards, research, patents, white papers, and additional languages.

🏝️ Benefits

• Remote work arrangement.

• Up to 25% travel for customer engagements, partner meetings, lab work, and industry events.

• Opportunities for collaboration within a distributed, international team.

• Chance to work directly with leading GPU cloud operators, NeoClouds, sovereign clouds, and AI-first enterprises.

• Opportunity to shape product narratives and influence go-to-market success.