Senior Network Engineer

Network Engineering Opportunities: GPU and AI Infrastructure

Several roles | [Sydney / Melbourne]

Our client is a fast-growing AI cloud provider building large-scale GPU compute infrastructure in Australia. They’re expanding their network team and hiring for several roles, from mid-level to senior and lead.

If you’ve worked on the network that sits under GPU clusters, or you want to, keep reading.

What you’ll be working on

• Designing, building and running high-performance, low-latency fabrics for large GPU clusters

• Back-end compute fabrics (InfiniBand and/or RoCE) and the front-end, storage and management networks around them

• Tuning RDMA performance for distributed AI training and inference

• Overlay and data centre networking with VXLAN/EVPN and BGP

• Troubleshooting hard problems at scale, including congestion, packet loss, latency and fabric health

• Automating builds and day-to-day operations so things stay repeatable as the platform grows

What we’re looking for

• Strong data centre networking experience in high-throughput environments

• Hands-on experience with RDMA and RoCE / converged Ethernet, including lossless Ethernet

• Solid VXLAN / EVPN and BGP experience in leaf-spine fabrics

• Experience in or around GPU infrastructure such as HPC, AI/ML clusters, cloud providers or research computing highly advantageous

• InfiniBand is a big bonus, but you don’t need it. Strong RoCE and Ethernet fabric people are just as welcome.

• Nice to have: NVIDIA networking, Arista or Cisco Nexus, Linux, and automation with Python or Ansible

Why this one

• Work on GPU infrastructure at scale

• A growing business where your work will be seen

• Several roles open, so there’s room to find the right level and focus for you

• Hybrid work in Sydney or Melbourne, options for remote too