Platform Engineer - Industrial AI Cloud

About the role

Deutsche Telekom IT Solutions is looking for a Platform Engineer to build, automate and operate the platform services of the Industrial AI Cloud, a joint project between NVIDIA and Deutsche Telekom for European manufacturers. The role centres on running and evolving large-scale Kubernetes clusters, container orchestration, CI/CD pipelines, GitOps and automation for AI workloads, and involves collaboration with infrastructure, security and AI teams.

Key responsibilities

• Build, configure and maintain bare metal hosts and Kubernetes clusters for GPU/AI workloads, and act as the main contact for Kubernetes issues, performance tuning and scaling.
• Design and operate the NVIDIA AI software stack (Slurm, Run AI) and provide customised application support for AI workloads.
• Manage Helm charts, GitOps workflows, Ansible scripts and possibly Terraform code to deploy services and AI workloads.
• Develop and maintain CI/CD pipelines with Jenkins and GitLab, and implement GitOps practices.
• Operate and improve Prometheus and Grafana monitoring stacks for hosts, Kubernetes and platform services.
• Build, optimise and secure container images (Docker, Podman), manage registries and run image scanning with Trivy.
• Maintain object storage and persistent volumes, and support distributed AI and HPC workloads.
• Follow ITIL incident, problem and change management workflows, maintain runbooks and adhere to zero-outage guidelines.

Required profile

• Kubernetes Certified Administrator (CKA) or equivalent production experience; CKS is an advantage.
• Strong experience with CI/CD tools (Jenkins, GitLab) and GitOps.
• Proficiency with Helm charts and Kubernetes resource management.
• Scripting in Python or Bash, and Infrastructure-as-Code with Terraform and Ansible.
• Experience with container images, object storage, monitoring tools and running AI/HPC workloads at scale.
• Preferred: ability to automate repetitive tasks and build self-service capabilities for developers, and a security-conscious approach.

Required skills

• Kubernetes (CKA/CKS)
• Helm, Docker/Podman, GitOps, Jenkins, GitLab
• Terraform, Ansible, Python or Bash scripting
• Prometheus, Grafana, Trivy
• NVIDIA GPU-accelerated server platforms and the NVIDIA AI software stack

What we offer

• Work on Europe's first industrial AI cloud with NVIDIA and Deutsche Telekom experts.
• Hybrid working model, training opportunities and career progression.