Site Reliability Engineer (SRE) - K8S/Slurm

GMI Cloud is an AI-native infrastructure company delivering high-performance GPU compute, inference services, and infrastructure for AI agents. The Site Reliability Engineer will ensure the stability, efficiency, and reliability of large-scale AI/ML clusters by designing infrastructure solutions, automating operations, monitoring systems, managing GPU node lifecycles, and responding to infrastructure incidents and customer provisioning requests.

Responsibilities:

• Design, implement and maintain scalable AI/ML infrastructure solutions
• Proactively monitor GPU cluster health, performance and troubleshoot issues across compute, accelerator, and storage systems
• Automate deployment, configuration and management of infrastructure resources
• Manage GPU node lifecycle workflows, including provisioning, scaling, maintenance, decommissioning and upgrades of GPU nodes
• Implement CI/CD pipelines for infrastructure deployment and orchestration
• Ensure security, compliance and best practices across infrastructure
• Manage incident response related to Infrastructure resources (GPU, CPU, Storage, Network)
• Handle customer provisioning requests for GPU resources, including onboarding, configuration and troubleshooting; resolve customer service requests related to GPU infrastructure, ensuring high customer satisfaction
• Stay current with emerging GPU hardware and software technologies, integrating improvements as appropriate
• Regional/international travel to GMI data center locations

Requirements:

• Bachelor's degree in Computer Science or related field
• Over 3+ years of experience in data center operations, infrastructure, or systems engineering
• Proven experience in site reliability engineering and infrastructure automation (e.g. Ansible, Terraform)
• Familiarity with containers orchestration platform (e.g. Kubernetes, Nvidia GPU operator, Nvidia Network operator, CNI, CSI) and job scheduling systems (e.g. Slurm)
• Familiarity with Linux system administration and scripting (Python, Bash)
• Familiarity with logging and monitoring tools such as Prometheus, Grafana, Loki
• Strong troubleshooting skills and ability to analyze system logs and performance metrics
• Excellent communication and teamwork abilities
• Good knowledge of GPU architecture, Nvidia CUDA, NCCL, or related AI/ML frameworks - added advantage