Domain Architect – Storage

12 Month Contract (3 year program of work)

Sydney OR Melbourne (Heavily remote, need to be okay with occasional site visits)

Day Rate: $1,400 - $1,600 (DOE)

About the prospect:

A major System Integrator building large scale, purpose built AI infrastructure for enterprise clients is looking for a Domain Architect to take ownership of the storage layer across their NVIDIA based deployments. It's a genuinely hands on role, split roughly 60% delivery and 40% pre sales support, spanning SuperPOD, BasePOD and equivalent AI factory builds. You'll be the person as comfortable configuring an NVMe over Fabrics connection in the CLI as explaining it to a C level client.

What you'll be doing:

- Leading installation and configuration of high performance storage clusters using platforms such as WEKA, DDN/Lustre, VAST Data or Pure Storage
- Implementing GPUDirect Storage to enable direct data paths between NVMe drives and GPU memory, bypassing the CPU
- Deploying CSI drivers for Kubernetes/Red Hat OpenShift and designing storage classes that separate high performance scratch tiers from general purpose home/project storage
- Running benchmark suites (IOR, FIO, mdtest) to validate throughput and metadata performance against agreed SLAs, and troubleshooting cases where slow I/O is starving GPU utilisation
- Implementing automated data tiering strategies across hot,



warm and cold storage based on access patterns
- Leading architectural sizing for storage opportunities, shifting the conversation from raw capacity to throughput per GPU
- Owning technical accuracy of the storage components list, validating interoperability against the NVIDIA HCL, and advising clients on the move from legacy NAS protocols to AI native storage

What we need to see:

- Expert level knowledge of parallel file systems such as WEKA, Lustre, BeeGFS or GPFS, and solid understanding of object storage for model checkpointing and archiving
- Mastery of storage protocols including NVMe over Fabrics, NFS over RDMA and GPUDirect Storage
- Deep understanding of the Linux I/O stack, including block device drivers, file system tuning (xfs, ext4) and client side caching
- Proficiency in Python, Ansible or Terraform for automating storage cluster deployment and configuration management

Bonus points:

- Background within a System Integrator, storage vendor or Managed Service Provider environment
- Hands on experience integrating storage with bare metal and Kubernetes/Red Hat OpenShift platforms
- Understanding of InfiniBand and RoCEv2 fabrics from a storage perspective, including congestion control and quality of service
- Exposure to LLM inference and caching tooling such as NVIDIA Dynamo, vLLM, SGLang or LMCache

📌 Domain Architect (Australia)
🏢 CircuIT Recruitment Group
📍 Australia