Senior Infrastructure Engineer, OpenStack

NexGen Cloud

full-time

Posted on: 9/22/2025

Origin: • 🇦🇺 Australia

✨ AI Apply

Senior

AnsibleCloudGrafanaKubernetesLinuxOpenStackPrometheusPythonTerraform

About the role

Design GPU-optimized Kubernetes clusters and build multi-tenant GPU infrastructure as part of GPUaaS offering.
Architect and deploy OpenStack and Kubernetes clusters designed for GPU scheduling, high performance, and multi-tenant workloads.
Automate deployment pipelines for cloud infrastructure using Terraform, Ansible, Helm, and Kubernetes Operators.
Build and manage GPU-ready container runtimes, NVIDIA device plugins, and Kubernetes-native GPU provisioning frameworks.
Ensure high availability and performance of OpenStack and Kubernetes clusters using tools such as Prometheus, Grafana, Loki, and Thanos.
Implement secure namespace isolation, RBAC, and network policies across OpenStack and Kubernetes layers.
Collaborate cross-functionally with DevOps, AI, Support, and Product teams to align infrastructure services with platform goals.
Contribute to automation, observability, and CI/CD tooling across the platform and provide guidance on Kubernetes best practices.

5+ years of experience with OpenStack in production environments.
3+ years of experience managing production-grade Kubernetes clusters, including bare-metal or private cloud environments.
Strong hands-on expertise with Kubernetes operators, Helm, and custom resource definitions (CRDs).
Experience with GPU orchestration in Kubernetes using NVIDIA tools.
Experience with multi-cluster or federated Kubernetes.
Proficiency in Linux, Ceph, networking (Calico/Cilium), and infrastructure scripting (Python, Bash).
Strong knowledge of cloud-native security, policy frameworks, and service meshes.
Experience with CI/CD pipelines, GitOps, and infrastructure-as-code tooling (Terraform, Ansible, ArgoCD).
Deep OpenStack and strong Kubernetes expertise.
Good to have: Experience integrating Kubernetes with OpenStack.
Good to have: Prior contributions to Kubernetes SIGs or CNCF projects.
Good to have: Knowledge of GPU metering, billing, and quota enforcement.
Good to have: Familiarity with HPC environments, InfiniBand/ROCEv2 networking, or Slurm integration.