GPU MLOps Engineer

Sourced from diversityjobs.com

Molex is hiring a GPU MLOps Engineer to own the Azure platform layer end-to-end for its engineering AI/ML platform. The role covers deploying AI/ML models, provisioning and scaling GPU compute, securing the environment, and continuously optimizing cloud cost. You will partner closely with data scientists, ML engineers, and LLM engineers on the compute, security, and cost backbone that powers their tooling.

• Build and maintain CI/CD pipelines in Azure DevOps for training, validation, versioning, and deployment of ML models
• Provision and scale Azure GPU compute (ND/NC series) and container infrastructure (AKS/ACI) for training and simulation workloads
• Automate retraining and redeployment workflows in Azure ML pipelines as new data becomes available
• Implement access control, identity management, and secrets management (Microsoft Entra ID, Azure Key Vault) across compute, data, and model artifacts
• Monitor and optimize GPU/cloud spend with Azure Cost Management and build cost visibility dashboards in Power BI

• 10+ years in MLOps, DevOps, or cloud infrastructure with exposure to both ML deployment and cloud infrastructure management
• Strong Azure experience including GPU compute (ND/NC), Azure Kubernetes Service, and containerization with Docker
• Infrastructure-as-code experience with Terraform or Bicep
• Solid grasp of cloud security fundamentals (IAM, network security, secrets management)
• Demonstrated cloud cost optimization experience (right-sizing, autoscaling, spot/preemptible strategies)
• Hands-on experience with model registries and experiment tracking (Azure ML + MLflow)

• Experience with GPU-heavy workloads such as ML training or HPC simulation
• Azure certifications (Solutions Architect Expert, Security Engineer Associate)
• Experience with HPC job schedulers such as Slurm via Azure CycleCloud
• Experience supporting engineering simulation tools