About the Role

One of our clients are building out their network backbone that connects and orchestrates thousands of GPUs at scale.

You'll be joining a team working with cutting-edge Nvidia hardware and high-performance networking technologies that few local organisations operate at this scale.

The role:

- Design, build, and operate high-performance network fabrics supporting large-scale GPU clusters
- Deploy and tune switch fabrics for low-latency, high-throughput AI/ML workloads
- Work closely with infrastructure and platform teams to optimise network performance for distributed training and inference
- Troubleshoot and resolve complex network issues across HPC and GPU-dense environments
- Plan capacity and topology for ongoing GPU infrastructure expansion
- Partner with vendors (including Nvidia) on hardware qualification, firmware, and fabric design
- Contribute to standards and best practices for network architecture as the workplace scales

Experience needed:





- Hands-on experience with high performance computing (HPC) networking environments
- Strong understanding of GPU infrastructure and the networking demands of AI/ML workloads
- Experience with switch fabric design, deployment, and operations at scale
- Familiarity with Nvidia networking technologies (e.g. InfiniBand, Spectrum-X, NVLink/NVSwitch ecosystems)
- We'll also consider candidates without direct HPC/AI experience if they bring:
- Large-scale network engineering experience from a hyperscaler or major cloud provider (e.g. AWS, Azure, Google Cloud) or equivalent big-tech environment
- Proven experience designing or operating large, complex switch fabrics in production — this background is scarce in the Australian market and highly valued
- A track record of operating at scale (thousands of nodes/ports) rather than traditional enterprise networking

📌 Network Engineer - Compute / HPC (Sydney)
🏢 Pathway Search
📍 Sydney