Senior ML Infrastructure Engineer

About the Company

Orion Innovation is a premier, award-winning, global business and technology services firm. Orion delivers game-changing business transformation and product development rooted in digital strategy, experience design, and engineering.

Responsibilities

• Own the end-to-end infrastructure layer including GPU cluster configuration and CUDA runtime management
• Manage Kubernetes job orchestration and model serving
• Design and implement Kubernetes autoscaling policies and multi-node-pool designs
• Debug complex issues such as CUDA out-of-memory errors
• Manage infrastructure via IaC and Helm charts

Requirements

• Expert knowledge of Kubernetes and Azure Kubernetes Service (AKS)
• Experience with GPU node pools (NC/ND series), taints, and tolerations
• Proficiency with KEDA, device plugins, and driver compatibility
• Experience with CUDA, cuDNN, and PyTorch GPU inference
• Strong skills in Python for production environments
• Experience with Hugging Face Transformers and BERT-class models
• Proficiency with Docker, Helm, and IaC tools like Terraform or Bicep
• Knowledge of Azure infrastructure including VNet, Key Vault, and ADLS