AI Infrastructure Operations Engineer

About the role

The Global AI Infrastructure team enables resilient, high-performance compute environments for strategic clients across cloud, on-premises, and hybrid deployments. We design, build, and operate large-scale GPU and accelerated-computing infrastructure that supports demanding AI training and inference, simulation, and high-performance compute workloads. Our work spans strategy, architecture, modernization, operations, governance, and continuous improvement across the infrastructure stack. We build reusable operational tools, automation workflows, and platform capabilities that make repeatable infrastructure tasks safer, faster, and more scalable. We collaborate across the technology ecosystem to harness new capabilities, drive business transformation, and deliver dependable services at scale.

Responsibilities

• Design and implement accelerated-computing infrastructure solutions aligned to system architecture, deployment roadmaps, performance, scalability, resiliency, and governance requirements.
• Deploy, configure, and operate GPU-based clusters across bare-metal and containerized environments, using workload schedulers and Kubernetes orchestration to support AI training, inference, and high-performance compute workloads.
• Integrate infrastructure platforms with enterprise systems, data platforms, security frameworks, service-management processes, and governance controls.
• Design, build, and maintain reusable tools, scripts, self-service capabilities, and automation workflows for infrastructure operations, including provisioning, configuration management, validation, capacity planning, monitoring, incident management, reporting, and recurring remediation.
• Establish repeatable operational processes for cluster provisioning, configuration management, patching, capacity planning, monitoring, incident response, and lifecycle management.
• Perform and automate GPU, compute, storage, and network benchmarking and validation; diagnose performance issues across multi-node AI training, inference, and distributed compute workloads.
• Develop and maintain architecture diagrams, configuration baselines, operational runbooks, and support documentation.
• Provide technical guidance, troubleshooting, and optimization for GPU clusters supporting AI training, inference, high-performance computing, and multi-node simulation workloads, with emphasis on availability, resiliency, scalability, energy efficiency, and cost management.
• Travel may be required for this role. The amount of travel will vary from 25% to 60% depending on business need and client requirements.

Required Skills and Qualifications

• Minimum of 5+ years of experience designing, deploying, and managing accelerated-computing infrastructure across on-premises, cloud, and hybrid environments for hyperscaler, neocloud, large enterprise, telecommunications, financial services, manufacturing, and/or retail clients.
• Minimum of 5+ years of hands-on experience with accelerated-computing platforms, including GPUs, DPUs, and CPUs, high-bandwidth network fabrics, and AI based storage architectures such as parallel file systems, NVMe-oF, etc.
• Minimum of 5+ years of experience with cluster management, workload scheduling, orchestration, observability, and infrastructure automation