DevOps / AI Infrastructure Engineer - GPU & Kubernetes

Senior DevOps / AI Infrastructure Engineer to design, build and operate GPU-accelerated AI/ML infrastructure. You will enable high-performance training and inference workflows by managing cloud/GPU platforms, Kubernetes clusters, IaC, and AI tooling (Triton, Kubeflow, MLflow). The role combines deep platform engineering with automation and close collaboration with ML engineers and R&D teams., * Design, deploy and operate GPU-enabled Kubernetes clusters and associated platform services for training and inference.

• Build and maintain CI/CD, model CI and MLOps pipelines using tools such as Kubeflow, MLflow and Triton.
• Implement and manage cloud infrastructure on Azure (and other clouds as needed), with GPU instances and storage for large datasets.
• Automate provisioning and configuration using Terraform, Ansible and scripting (Python, Bash).
• Optimize container orchestration, scheduling and GPU utilization for high-performance workloads.
• Integrate AI inference platforms (NVIDIA Triton) and support model serving at scale.
• Work with PLM/simulation and data teams to integrate model training and inference into engineering workflows.
• Monitor, troubleshoot and tune platform performance, reliability and cost.
• Define and enforce best practices for security, resource governance and data handling in AI pipelines.
• Document architectures, runbooks and operational procedures; transfer knowledge to engineering teams.

Start: ASAP (or as agreed) Duration: 6 months (with possibility to extend) Experience: 8-10 years (including at least 1.5 years in DevOps/cloud/SRE focused on AI/ML) Language: English (fluent), * 8-10 years industry experience; minimum 1.5 years in DevOps, cloud engineering or SRE with AI/ML focus.

• Hands-on experience with major cloud providers (Azure preferred; experience with GCP/AWS is valuable).
• Experience with GPU-accelerated environments and NVIDIA ecosystem.
• Deep understanding of Kubernetes and container orchestration for high-performance computing and model serving.
• Experience with AI platform tools: NVIDIA Triton, Kubeflow, MLflow (setup, pipelines, serving).
• Strong scripting and programming skills (Python, Bash) for automation and data processing.
• Proficiency with Infrastructure-as-Code: Terraform and configuration management with Ansible.
• Solid knowledge of storage, networking and security considerations for large ML workloads.
• Good communication and collaboration skills; able to work with ML researchers and engineering teams.

Preferred skills

• Experience integrating AI/ML workflows with PLM or simulation platforms.
• Familiarity with GPU scheduling solutions (NVIDIA GPU Operator, device plugins, Volcano, etc.).
• Knowledge of monitoring/observability for ML platforms (Prometheus, Grafana, ELK, metrics for GPU workloads).
• Experience with cost-optimisation and autoscaling strategies for GPU clusters.
• Familiarity with model optimization techniques (quantization, batching, mixed precision) and inference performance tuning.

What we offer

• Work on cutting-edge AI infrastructure supporting R&D and engineering use cases.
• Opportunity to shape GPU and MLOps practices in a collaborative technical environment.
• Competitive compensation and Eindhoven-based role with flexible working arrangements.