Infrastructure Engineer, Hardware & Networks

Entangled Future; Sydney; Hybrid; AUD 120,000 - 180,000; Full time

Job summary

Silicon Quantum Computing is hiring an Infrastructure Engineer in Sydney for Platform & Infrastructure. The role focuses on the physical and network end of the platform, with on-premise compute, clusters, FPGA hosts and Kubernetes above all of it.

You will rack and wire hardware, implement declarative network configurations, and automate provisioning from bare metal to a secured Kubernetes node. This position is based at the Sydney facility and involves collaboration with IT, facilities and

Qualifications

• 4+ years in infrastructure, network or platform engineering with bare-metal work.
• Deep Ansible experience: roles, inventories, idempotent design and testing.
• Solid network engineering across L2/L3: VLANs, routing, ACLs.

Responsibilities

• Own bare-metal cluster server fleets end to end: racking, cabling, provisioning and lifecycle.
• Design, build and operate network infrastructure with central IT: VLANs, routing, firewall policies.
• Drive all of it through Ansible and infrastructure-as-code for repeatable builds.
• Build automated provisioning pipelines from power-on to ready cluster.
• Run and maintain Kubernetes on hardware, including CNI, storage and node lifecycle.
• Operate core services: DNS, DHCP, NTP, PXE, certificates.

Skills

Infra engineering

Ansible

Network engineering

Tools

Kubernetes

Job description

Infrastructure Engineer, Hardware & Networks at Silicon Quantum Computing, Sydney (remote). A mid-level full time position in the Software team, listed under hardware engineering.

About the role

We are hiring an Infrastructure Engineer into Platform & Infrastructure, at the physical and network end of the platform, driven from code: the servers, racks, switches and fabric everything else in the company runs on.

Most of our compute is on-premise and growing. Simulation clusters, FPGA accelerator hosts for the quantum control stack, lab and instrument networks, and Kubernetes above all of it. When a team needs a cluster, someone racks it, cables it, wires it into the fabric, configures it in code, gives it addresses, names and time, and hands it over reproducible. Every quantum computer we build also gets a cluster of its own, so the count tracks machines rather than headcount. Soon more than ten edge clusters will exist across the R&D estate.

We run these fleets in code: Ansible roles, version-controlled inventories, declarative network configuration, and provisioning that goes from bare metal to a secured Kubernetes node without anyone typing into a terminal. Some of what you connect is a general-purpose compute node. Some is a control host with accelerator cards in it, metres from a dilution refrigerator, with timing requirements ordinary enterprise networks never have.

Based at our Sydney facility, you will work with central IT, facilities and the research and engineering teams whose hardware you stand up. This is a role for someone equally at home in a comms room and a Git repository, who wants every build to be repeatable and every change reviewable.

Essential

• 4+ years in infrastructure, network or platform engineering, with substantial hands-on bare-metal work
• Deep Ansible experience: roles, inventories, idempotent design, and how to test and structure a real playbook estate
• Solid network engineering across layers 2 and 3: VLANs, subnetting, routing protocols such as BGP or OSPF, link aggregation, firewall and ACL policies

Responsibilities

• Own the bare-metal cluster server fleets end to end: racking, cabling, provisioning, firmware and BMC management, lifecycle and decommissioning, across central, research and control clusters
• Design, build and operate our network infrastructure together with central IT: switching, routing, VLANs and segmentation, link and route aggregation, firewall policy, and the high-speed fabrics behind the compute and storage
• Drive all of it through Ansible and infrastructure-as-code, so builds are repeatable and changes are reviewable
• Build and maintain the automated provisioning pipelines that take a new rack from power-on to a validated, benchmarked, ready cluster
• Build the cluster that ships with each quantum computer, to the same specification every time, in our own labs
• Run and maintain Kubernetes on this hardware, including CNI, load balancing, storage integration and node lifecycle
• Operate core network services: DNS, DHCP, NTP and time distribution, PXE and netboot, certificates
• Support hardware bring-up: BIOS and firmware tuning, BMC configuration, PCIe topology and NUMA alignment on accelerator hosts
• Support the compute, control and lab hardware that research and engineering teams depend on, including accelerator-card hosts and instrument networks
• Instrument the servers, fabric and provisioning path so our monitoring and telemetry function has something to work with

Nice to Have

• Low-latency or high-throughput fabrics: 100/400G Ethernet, InfiniBand, RoCE
• EKS Anywhere experience
• Precision time distribution (PTP, White Rabbit or similar)
• Network automation and source-of-truth tooling (NetBox, Nornir, containerlab, Batfish)
• Vendor platforms such as Arista EOS, Cisco NX-OS, Juniper Junos, Dell OS10
• HPC, scientific or research computing environments, including batch schedulers such as Slurm
• Cluster benchmarking and performance validation, and the automation around it
• Monitoring and telemetry (Prometheus, Grafana, syslog, Filebeat)