Senior AI Infrastructure & Platform Engineer

Deepsource Technologies

Key Responsibilities

• Deploy, maintain, and optimize GPU-based compute clusters and infrastructure.
• Manage and operate GPU orchestration tools and platforms such as:
• Nvidia Base Command Manager
• Nvidia AI Enterprise Suite
• Nvidia GPU and Network Operators
• Nvidia NIMs and Blueprints
• Configure, deploy, and maintain compute workloads using scheduling and orchestration tools including:
• Slurm
• Vanilla Kubernetes
• Install, configure, and maintain the underlying OS (e.g. Canonical Ubuntu) and supporting system software.
• Monitor and troubleshoot infrastructure performance, availability, and reliability; ensure high uptime for AI/ML workloads.
• Work with data scientists, ML engineers, and dev teams to define infrastructure requirements, resource allocation, and deployment workflows.
• Develop automation scripts, CI/CD pipelines, and best practices for infrastructure provisioning and management.
• Document architecture, configurations, and operational procedures; enforce security, compliance, and backup policies.

Required Skills & Experience

• Proven experience managing GPU-based AI/ML infrastructure and compute clusters.
• Hands-on experience with:
• Nvidia Base Command Manager
• Nvidia AI Enterprise Suite
• Nvidia GPU/Network Operators, NIMs, Blueprints
• Strong experience with Slurm and/or Kubernetes orchestration.
• Solid Linux system administration skills preferably on Ubuntu or similar distributions.
• Strong scripting/automation ability (e.g. Bash, Python, or relevant tooling) for provisioning, deployment, and maintenance.
• Excellent troubleshooting and performance-tuning skills.
• Experience collaborating with ML/data science teams and integrating infrastructure with their workflows.
• Strong understanding of networking, security, resource allocation, and cluster management best practices.

Preferred Qualifications

• Previous experience working in a high-performance computing (HPC) or AI-focused infrastructure team.
• Knowledge of containerization, container orchestration, and GPUs in cloud or on-prem environments.
• Experience with CI/CD, infrastructure-as-code (e.g. Terraform, Ansible), monitoring tools, and logging setups.
• Familiarity with workload scheduling, job queuing, resource quotas, and GPU-shared environments.