We are seeking an experienced HPC Systems Engineer to design and manage the high-performance computing infrastructure supporting our AI workloads — from deep learning training to large-scale data processing for educational analytics.

Responsibilities

• Architect and maintain HPC clusters and GPU-based compute environments for ML model training.
• Optimize distributed training pipelines (PyTorch DDP, Horovod, or Ray).
• Manage job scheduling systems (Slurm, Kubernetes, or Azure Batch) for efficient workload allocation.
• Ensure scalable data access and high-throughput pipelines between storage and compute nodes.
• Collaborate with AI/ML and MLOps teams to integrate training pipelines with CI/CD workflows.
• Implement security, resource monitoring, and cost optimization practices for compute clusters.
• Automate cluster provisioning and teardown to support on-demand AI workloads.

Requirements

• 5+ years managing HPC or GPU-based compute infrastructure.
• Proficiency with Linux, Slurm, Kubernetes, Docker, and Terraform.
• Hands-on experience with Azure Batch, AWS ParallelCluster, or on-prem HPC setups.
• Strong understanding of networking, parallel file systems, and GPU performance tuning.
• Familiarity with data pipelines and model training orchestration.
• Excellent scripting skills (Bash, Python).