Senior Solutions Architect – AI Infrastructure Networking (remote on the east coast)

Description

You will join the Voyager team in the office of the CTO as a senior solutions architect focused on networking.

You will own the network design for the company's AI infrastructure platform, which builds and runs GPU clouds and AI factories from bare metal up to models.

The role mixes research with hands-on, customer-facing work. You will explore designs, write architecture, validate it in the lab and explain it to engineers, executives and conference audiences.

Research is central. You will look at designs before customers ask for them, prototype ideas that may never ship, and question accepted practice when something better exists. Your work will shape the platform's future as well as its current deployments.

You will work with solutions architects, partner management and engineering colleagues across many countries and time zones.

The company is a Kubernetes-focused AI infrastructure business serving platform engineering teams across on-premises, cloud, edge and sovereign data centers.

Responsibilities

• Design and publish network reference architectures for data center layouts from a single rack to clusters with thousands of GPUs.
• Specify east-west compute fabrics on InfiniBand and RoCEv2 Ethernet, including rail-optimised, fat-tree and Clos designs, and choose oversubscription and failure domains.
• Design front-end, storage and management networks and how they link 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 covering cost, performance, operability and vendor lock-in, with reasoning customers can follow.
• Define how Linux hosts expose NICs, DPUs and SuperNICs to workloads, including PCI passthrough, SR-IOV virtual functions, IOMMU groups, NUMA and GPU-NIC affinity.
• Design Kubernetes networking for AI workloads: primary and secondary CNIs, Multus, SR-IOV and RDMA device plugins, the NVIDIA Network Operator and Dynamic Resource Allocation (DRA).
• Design VM networking on KubeVirt, including passthrough and SR-IOV for GPU and RDMA traffic.
• Track and evaluate emerging AI networking technologies and standards, such as Ultra Ethernet (UEC), 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 evidence supports a change, 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 on lab hardware and customer sites, and report results against agreed success criteria.
• Write automation and tooling in Python, Go, Bash, Ansible, Helm, Kubernetes manifests and Terraform so the reference architectures can be reproduced.
• Run validation and benchmarks on fabrics and host setups using tools such as NCCL tests, perftest and ibwritebw, and turn results into design guidance.
• Act as the network expert in customer discovery, design reviews and architecture workshops, bringing new ideas rather than only restating current practice.
• Work with hardware, networking and switch partners such as NVIDIA, server OEMs and network vendors on joint designs and validation.
• Feed research results back to product and engineering and help shape the platform roadmap.
• Present at industry events, webinars and partner summits, and lead hands-on workshops for technical audiences.
• Write technical content: reference architectures, solution briefs, blog posts and internal enablement material.

Requirements

• Bachelor's degree in computer science, electrical engineering, telecommunications or a related field, or equivalent practical experience.
• 8+ years in network engineering or 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 challenges of running 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, VPNs, 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, with 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.
• Virtualisation on Kubernetes: KubeVirt, including how VM networking, passthrough and SR-IOV are configured.
• Programming or scripting in at least one language, Python or Go preferred, and comfort with Git, CI and infrastructure as code.
• Behaviours, with concrete examples of each requested: curiosity about why fabrics behave as they do; ownership from first sketch to validated result; proactive identification of gaps; fast learning in unfamiliar environments; and business sense linking design choices to cost, time-to-value and risk.
• Excellent written and spoken English, with the ability to write documents others can build from, present to large audiences, run workshops and adapt the message from engineers to executives.
• Experience in an international, distributed company across time zones and work cultures, working with solutions architects, partner management and engineering teams.
• Nice to have: hands-on NVIDIA networking, including Quantum InfiniBand, Spectrum-X Ethernet, ConnectX SuperNICs, BlueField DPUs, UFM and NVIDIA AI factory reference designs.
• Nice to have: experience with GPU cloud, neocloud or HPC operators, or NVIDIA Cloud Partner (NCP) style deployments.
• Nice to have: bare-metal provisioning and lifecycle with Metal3, Ironic, Redfish, PXE or iPXE, and Cluster API.
• Nice to have: network automation, SDN controllers, and switch operating systems such as Cumulus Linux, SONiC or equivalents.
• Nice to have: global load balancing and service exposure for distributed inference, including GSLB, Gateway API and ingress.
• Nice to have: software-defined storage networking for AI, including NVMe-oF and GPUDirect Storage.
• Nice to have: experience with the company's own Kubernetes products, MKE, k0s and k0rdent networking.
• Nice to have: contributions to open-source networking or Kubernetes projects, or a record of public talks and technical writing.
• Nice to have: participation in standards bodies such as the Ultra Ethernet Consortium, OCP or IETF, or published research, patents or white papers.
• Nice to have: languages beyond English, such as a European language.

Conditions

This role is remote and open to candidates based in 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.

The company is an equal opportunity employer.

Benefits

• Pay and benefits are set according to the local market and your employment arrangement.
• Work directly with GPU cloud operators, NeoClouds, sovereign clouds and AI-first enterprises.
• Work alongside a distributed team committed to openness and technical excellence.
• Influence the product story and go-to-market results.