Senior AI Platform Engineer

We are ResetData. And we’re on a mission to offer sustainable cloud infrastructure solutions that are better for businesses and the planet.

ResetData significantly reduces the energy and water resources required to run data centres. Our liquid cooling technology also enables a tenfold increase in computer density, equating to significant cost savings to businesses seeking smarter, more sustainable solutions. We are also a leading public cloud provider, specializing in cutting-edge GPU as a Service (GPUaaS) for AI and Machine Learning workloads.

With a seasoned executive team, backing from Centuria Group, and partnerships with premier technology providers, ResetData is making an enduring environmental difference, redefining a saturated market and legacy for generations to come.

This is your chance to join as part of ResetData’s foundational team as we shape the future of next generation cloud services.

Location: Sydney or Melbourne

About the Role

As a Senior AI Platform Engineer, you will own the build and day-to-day operations of GPU server environments across our AI Factory infrastructure. Your scope spans the full stack - from bare-metal server provisioning and GPU hardware activation, through BCM-managed clusters and Kubernetes platforms, to the automated deployment and service-delivery layer that underpins our customer-facing offerings (BMaaS, GPUaaS, Managed Clusters, and LLMaaS).

You will work at the intersection of high-density hardware engineering and cloud-native platform operations, ensuring that every layer of the stack - silicon, firmware, networking, orchestration, and AI services - is performant, stable, and production-ready. You will be a key contributor within a small, collaborative team building and scaling a next-generation AI compute platform.

The role is distinguished from a general systems engineer role by its deep focus on GPU hardware lifecycle, NVIDIA software stack proficiency, and the ability to bridge bare-metal provisioning through to AI/ML service delivery.

What You'll Do

Hardware & Bare-Metal

• Provision, operate and maintain bare-metal GPU servers across our NVIDIA fleet (L40S, RTX 6000, H200, B300, GB300 NVL72). This covers firmware, monitoring, fault diagnosis and power budgeting (DPS/MaxLPS).
• Run out-of-band management (IPMI/BMC, Redfish) for remote provisioning, health monitoring and headless deployments.
• Support the Systems Engineers on fleet lifecycle management through Dell OpenManage Enterprise.
• Support the Network Engineers on the InfiniBand/Spectrum-X fabric, BlueField DPUs, RDMA/RoCE v2 and customer multi-tenancy.

GPU Performance

• Tune and troubleshoot CUDA, NCCL and UCX for multi-node training and inference, including transport selection and topology-aware placement.
• Validate performance using DCGM, nvidia-smi and structured L11 (rack) and L12 (cluster) acceptance testing.
• Find and fix bottlenecks, thermal issues and GPU degradation before customers notice them.

Platform & Orchestration

• Build and run GPU-optimised Kubernetes and OpenShift clusters on-prem, including the GPU Operator, container runtimes and GPU-aware scheduling.
• Administer NVIDIA Base Command Manager and the Rafay-based automated provisioning for multi-tenant infrastructure.
• Design and enforce multi-tenancy boundaries: RBAC/ABAC, network segmentation, encryption, TLS and secrets management.

Service Delivery & Reliability

• Support our customer-facing services (BMaaS, GPUaaS, Managed Clusters and LLMaaS), including inference and serving platforms where needed.
• Build automation and infrastructure-as-code (OpenTofu, Ansible, Helm) for provisioning, monitoring and incident response.
• Own observability for GPU and platform health using Prometheus, Grafana, DCGM Exporter, LibreNMS and SigNoz.

Collaboration

• Work closely with the Systems, Network, product and service design teams on AI Factory builds, feasibility of new services, runbooks and the infrastructure roadmap

What You’ll Bring

• You’re an Australian citizen (required for security clearances)
• 8+ years of Linux systems engineering in production data centre environments.
• Hands-on production experience with NVIDIA data centre GPUs, drivers, firmware, CUDA and basic NCCL tuning.
• Strong Kubernetes for AI/ML: cluster deployment, GPU Operator, GPU scheduling and container runtimes.
• Bare-metal provisioning and hardware lifecycle management in high-density compute.
• Solid high-performance networking: InfiniBand, 25–400GbE, RDMA/RoCE v2, TCP/IP, DNS and segmentation.
• Python for automation, plus infrastructure-as-code (OpenTofu/Terraform, Ansible) and GitOps.
• Hands-on observability experience (Prometheus, Grafana, DCGM or similar).
• A systematic way of troubleshooting across the hardware, OS, network and application layers.
• Comfortable in a start-up where infrastructure is being built, not just maintained.
• A clear communicator, from runbooks to customer-facing material.
• Collaborative, and genuinely into GPU computing and the NVIDIA ecosystem

What else you need to know:

• Flexible work environment: We value outcomes over hours. We work hybrid from our Sydney & Melbourne CBD offices, striking a balance between professional and personal commitments.
• Break new ground: help shape a company pioneering how AI Infrastructure in Australia is built & maintained.
• Opportunities for professional growth and development: we want you to hone your craft and learn new skills & new technologies.
• Team first: we take pride in our collaborative and inclusive team culture.

Sound like you? Come join us as we shape the future of sustainable cloud infrastructure.