Opportunity to join a market leading System Integrator delivering Multi-Billion-Dollar AI Data Centre Upgrade Programs. Work with the Recent and Emerging technology such as NVIDIA in an Org organisation at the forefront of delivering AI Data Centre Upgrade Programs, Delivering GPUs at scale. We have a (12-Month Contract - Permanent) Domain Architect - AI Compute roles available. You will join a specialist team delivering high-performance GPU and AI infrastructure for enterprise clients. This is a hands-on architecture and delivery role requiring strong NVIDIA GPU, HPC and Linux expertise. You will own the Compute layer across complex AI infrastructure deployments, working across architecture, cluster provisioning, automation, performance optimisation and Day 2 operations. Key Responsibilities:

Design and deliver

NVIDIA SuperPOD, BasePOD, NVL72, HGX, MGX and Cisco AI Factory

environments. Lead bare-metal GPU cluster provisioning using

NVIDIA Base Command Manager (BCM)

and Zero Touch Provisioning. Configure and optimise

Slurm, NVIDIA Run, Kueue and/or Volcano . Implement GPU monitoring using

NVIDIA DCGM

and validate performance using

NCCL, HPL and HPCG . Optimise Linux environments for high-performance AI/HPC workloads, including kernel, driver and system tuning.



Develop infrastructure automation using

Python and Ansible . Integrate compute environments with

InfiniBand/RoCEv2, storage, identity and enterprise platforms . Provide technical leadership to clients and support solution design, scoping and pre-sales activities. Essential Experience:

Strong HPC Architecture & Engineering Experience Strong Experience in Data Centre - Compute Strong hands-on experience with NVIDIA GPU/AI infrastructure and HPC environments . Deep knowledge of

NVIDIA Hopper, Grace Hopper, Blackwell and Grace Blackwell

platforms. Experience with

NVL72, NVSwitch, CUDA, cuDNN and NCCL . Advanced

Linux

systems engineering skills across Ubuntu and/or RHEL. Strong

Python and Ansible

automation experience. Understanding of GPU/PCIe/NUMA topology and high-performance compute architectures. Desirable Experience:

NVIDIA BCM/Mission Control, Slurm, Run, Kubernetes/OpenShift, InfiniBand NDR/HDR, RoCEv2, Rafay, KubeFlow, OpenShift AI, liquid cooling

and AWS/Azure/GCP/OCI. Experience within a

System Integrator, MSP, hyperscaler or specialist AI infrastructure environment

is highly regarded.

#J-18808-Ljbffr

📌 Domain Architect - Ai Compute - 12M Contract - Perm (Melbourne)
🏢 Northbridge Recruitment
📍 Melbourne