About XENON
XENON architects and builds AI factories — the dense, GPU-accelerated data centres that power large-scale training and inference. We stand up the compute, storage and networking fabric that AI workloads run on, end to end. As demand for GPU clusters accelerates, the network fabric has become the bottleneck that separates a fast cluster from a merely large one — and that’s the problem this role solves.
The role
We’re hiring a Senior Network Engineer to design, deploy and operate the high-performance fabrics connecting our GPU clusters. You’ll work directly on NVIDIA Cumulus Linux switching and InfiniBand fabrics, building the low-latency, lossless networks that GPU-to-GPU communication depends on — the difference between GPUs sitting idle and a cluster running at full utilisation.
What you’ll do
- Design and deploy leaf-spine fabrics for GPU clusters using NVIDIA Cumulus Linux and NVIDIA Spectrum switching.
- Build, tune and troubleshoot InfiniBand fabrics (subnet management, adaptive routing, congestion control) for large-scale AI training workloads.
- Own EVPN/VXLAN and BGP-based data centre routing designs, including automation of fabric provisioning and lifecycle.
- Work with RoCE/RDMA networking to optimise east-west GPU traffic and minimise tail latency.
- Partner with compute, storage and site reliability teams to plan capacity for new AI factory builds and expansions.
- Diagnose and resolve fabric-level performance issues affecting training job throughput — you’ll be the person GPU teams call when jobs run slow.
- Contribute to run books, monitoring and automation that let the fabric scale without scaling headcount 1:1.
What you bring
- Hands-on production experience with NVIDIA Cumulus Linux (or Cumulus Networks pre-acquisition) in a data centre environment.
- Direct experience designing, deploying, or operating InfiniBand fabrics — HPC, AI training clusters, or equivalent high-performance environments.
- Strong grounding in data centre routing: BGP, EVPN, VXLAN, leaf-spine architectures.
- Familiarity with RDMA/RoCE and the networking demands of GPU-to-GPU communication (NCCL-aware a strong plus).
- Comfort with network automation and infrastructure-as-code (Ansible, Python, NetDevOps tooling).
- Prior experience in HPC, hyperscale, or AI infrastructure environments is highly valued.
- NVIDIA InfiniBand Professional or NVIDIA-Certified Professional: AI Networking (NCP-AIN) certification is a plus, not a requirement.
Why XENON
- Work on the fabric layer of next-generation AI factories — not a legacy enterprise network.
- Direct exposure to the newest NVIDIA networking platforms (Cumulus Linux, Spectrum-X) as they ship.
- A small, senior team where your fabric designs go straight into production clusters.
- Competitive salary, plus the training and certification support to stay current on a fast-moving stack.
- Open to candidates based overseas — visa sponsorship and relocation support to Australia are available for the right person.
Diversity and Inclusion
XENON is committed to creating a diverse environment that embraces differences and fosters inclusion.
Equal Opportunity
XENON is an Equal Opportunity Employer and strongly committed to all policies which will afford equal opportunity employment to all qualified persons without regard to age, national origin, race, ethnicity, creed, gender, disability, veteran status, or any other characteristic protected by law.
Apply
Please complete the form below and tell us about the largest InfiniBand or Cumulus Linux fabric you’ve worked on, and tell us what broke and how you fixed it — that’s the story we care about most.