Senior Solutions Architect – AI Infrastructure Networking

Description

You would join the team in the Office of the CTO as a senior solutions architect with deep networking skills.

You would own the network design for the company's platform for building and running GPU clouds and AI factories, covering the stack from bare metal up to models.

Your designs would cover how large GPU clusters connect, how tenants stay separated, and how workloads reach the network from containers, virtual machines and bare metal nodes.

Your output would be published reference architectures, code and proofs of concept tested on real hardware with customers and partners.

Research is central to the role. You explore designs before customers ask for them, build prototypes that may never ship, and question accepted practice. Your work shapes the product's future direction, not only how it is deployed today.

You would split your time between research, writing architecture, building and validating in the lab, and presenting to engineers, executives and conference audiences. You would work with Solutions Architects, Partner Management and Engineering across countries and time zones.

The company is a Kubernetes-native AI infrastructure company that works with GPU cloud operators, NeoClouds, sovereign clouds and AI-first enterprises. It is part of a larger group.

Responsibilities

• Design and publish network reference architectures and solution designs, from a single rack up to clusters of thousands of GPUs.
• Specify east-west compute fabrics on InfiniBand and RoCEv2 Ethernet, including rail-optimised and fat-tree or Clos layouts, with decisions on oversubscription and failure domains.
• Design front-end, storage and management networks, and how they connect to customer data centers, public clouds and hybrid setups.
• Define multi-tenant isolation across the stack using InfiniBand PKeys, VRFs, EVPN-VXLAN and Kubernetes network policy.
• Document scale limits and trade-offs in cost, performance, operability and vendor lock-in, so customers can follow the reasoning.
• Define how Linux hosts expose NICs, DPUs and SuperNICs to workloads, covering PCI passthrough, SR-IOV virtual functions, IOMMU groups, NUMA and GPU-NIC affinity.
• Design Kubernetes networking for AI workloads, including primary and secondary CNIs, Multus, SR-IOV and RDMA device plugins, the NVIDIA Network Operator and Dynamic Resource Allocation.
• Design VM networking on KubeVirt, including passthrough and SR-IOV for GPU and RDMA traffic.
• Track and assess emerging AI networking work, such as Ultra Ethernet, SuperNIC and DPU offloads, scale-up and scale-across fabrics, and Kubernetes DRA network drivers.
• Prototype alternative designs in the lab, including ones no customer has requested, and benchmark them against current practice.
• Challenge established or customer-preferred designs when the evidence points elsewhere, and make the case with data.
• Publish findings as internal research notes and design proposals, and where suitable as external papers, blog posts or talks.
• Build and run proofs of concept with customers and partners in the company's labs and on customer sites, and report results against agreed success criteria.
• Write automation and tooling in Python, Go, Bash, Ansible, Helm, Kubernetes manifests or Terraform so the reference architectures can be reproduced.
• Run validation and benchmarks of fabrics and host configurations, such as NCCL tests, perftest and ibwritebw, and turn the results into design guidance.
• Act as the network expert in customer discovery, design reviews and architecture workshops, bringing new ideas as well as current practice.
• Work with NVIDIA, server OEMs, and switch and network vendors on joint designs and validation.
• Feed research results back to Product and Engineering and help shape the platform roadmap.
• Speak at industry events, webinars and partner summits, and run hands-on workshops for technical audiences.
• Write reference architecture documents, solution briefs, blog posts and internal enablement material.

Requirements

• A 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, including 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 design: spine-leaf and Clos topologies, rail-optimised GPU fabrics, oversubscription, ECMP, and the operational problems of scaling to thousands of nodes.
• InfiniBand: subnet management, partitioning, adaptive routing, and how NCCL and RDMA traffic behaves on the fabric.
• RoCEv2 on Ethernet: lossless, congestion-controlled designs using PFC, ECN and DCQCN, plus QoS, MTU and buffer tuning.
• Routing and overlays: BGP, EVPN-VXLAN and VRFs for multi-tenancy.
• Hybrid and multi-cloud connectivity: interconnects, VPN, cloud provider networking constructs, and IP address and DNS planning across sites.
• Linux networking: how the kernel and drivers manage NICs, PCI passthrough, SR-IOV, IOMMU and NUMA affinity, using tools such as iproute2, ethtool and devlink.
• Kubernetes networking: CNI plugins such as Cilium, Calico and OVN-Kubernetes, Multus and secondary networks, SR-IOV and RDMA device plugins, network policy and service exposure.
• KubeVirt, including how VM networking, passthrough and SR-IOV are configured.
• Programming or scripting in at least one language, preferably Python or Go, and comfort with Git, CI and infrastructure as code.
• The company assesses these behaviours as much as technical depth and will ask for concrete examples of each.
• Curious: you dig into why a fabric behaves as it does, read specs and source code, and keep up with how AI networking is changing. You question established designs, including ones customers are used to.
• Ownership: you take a reference architecture or proof of concept from first sketch to a published, validated result, and stand behind it.
• Proactive: you spot gaps in the product, documentation or a customer design, and act before you are asked.
• Fast learner: you become productive quickly on unfamiliar hardware, software or customer environments.
• Business acumen: you link network design choices to cost, time-to-value and risk, and explain them to decision makers in those terms.
• Excellent written and spoken English, and you write documents others can build from.
• Comfortable presenting to large audiences and running hands-on workshops.
• Able to adapt the message from network engineers to customer executives.
• Experience working in an international, distributed company across time zones and work cultures, with Solutions Architects, Partner Management and Engineering teams.
• Hands-on experience with NVIDIA networking: Quantum InfiniBand, Spectrum-X Ethernet, ConnectX SuperNICs, BlueField DPUs, UFM, and NVIDIA reference designs for AI factories.
• Experience with GPU cloud, neocloud or HPC operators, or with NVIDIA Cloud Partner style deployments.
• Bare-metal provisioning and lifecycle: Metal3, Ironic, Redfish, PXE, iPXE and Cluster API.
• Network automation and SDN controllers, and switch operating systems such as Cumulus Linux, SONiC or equivalents.
• Global load balancing and service exposure for distributed inference: GSLB, Gateway API and ingress.
• Software-defined storage networking for AI: NVMe-oF and GPUDirect Storage.
• Experience with the company's own products, including MKE, k0s and its networking stack.
• Contributions to open-source networking or Kubernetes projects, or a record of public talks and technical writing.
• Participation in standards or industry bodies such as the Ultra Ethernet Consortium, OCP or IETF, or published research, patents or white papers.

Conditions

This is a remote role, open to candidates based in the United States (East Coast) or Europe.

The team spans many time zones, so some meetings fall outside standard local hours.

Travel of up to 25% for customer engagements, partner meetings, lab work and industry events, mainly within the EU and US.

Pay and benefits are set according to the local market and the employment arrangement.

The company is an equal opportunity employer.