Senior GPU Interconnect & Fabric Reliability Engineer

Firmus Technologies Pty Ltd.; Sydney; Hybrid; AUD 150,000 - 190,000; Full time

Job summary

Firmus Technologies Pty Ltd. is seeking a Senior Platform Reliability Engineer, Fabric and Interconnect, to own the reliability of GPU-to-GPU interconnects, high-performance network fabrics and DPU-based host networking.

You will implement guarded automation, lead major incident response, and work with vendors at engineering level to achieve permanent fixes. The role requires deep technical expertise, strong communication skills, and hands-on automation in a 24/7 environment.

Qualifications

• 8+ years of production network or interconnect infrastructure experience.
• Expertise with GPU interconnect fabrics (NVLink/NVSwitch) and topologies.
• In-depth knowledge of InfiniBand, RoCE and related fabrics (Spectrum-X).
• Experience with DPU/SmartNIC host networking and offload paths.
• IaC and automation practices delivered via code and peer review.
• Ability to script for automation using Python/Go/Bash.

Responsibilities

• Ensure reliable operation, automation and continuous improvement of GPU interconnect and network fabrics.
• Build guarded automation and remediation tooling for fabric faults and AI clusters.
• Diagnose and tune performance across interconnect stacks from app to link level.
• Operate DPU-based host networking and manage driver/firmware compatibility.
• Coordinate firmware upgrades, capacity changes and canary rollout with rollback.
• Provide senior escalation support for fabric incidents and drive permanent fixes.
• Lead vendor escalations and mentor engineers with runbooks and procedures.

Skills

High-performance networking

24/7 production

GPU interconnect fabrics

NVLink/NVSwitch

InfiniBand / RoCE / Spectrum-X

DPU/SmartNIC host networking

Infrastructure automation / IaC

Scripting: Python/Go/Bash

Incident response & vendor escalations

Job description

Firmus Technologies Pty Ltd. is seeking a Senior Platform Reliability Engineer, Fabric and Interconnect, to own the reliability of GPU-to-GPU interconnects, high-performance network fabrics and DPU-based host networking.

You will implement guarded automation, lead major incident response, and work with vendors at engineering level to achieve permanent fixes. The role requires deep technical expertise, strong communication skills, and hands-on automation in a 24/7 environment.