Technical Program Manager – AI Infrastructure / GPU Clusters

Role overview

GPU cluster deployment

• Lead end-to-end deployment of AI GPU clusters from infrastructure planning through production launch.
• Manage timelines for hardware deployment, network integration, cluster bring-up, and production readiness.
• Coordinate infrastructure architects, network engineers, hardware vendors, and data center teams.

Architecture and integration

• Collaborate on GPU server selection, distributed-cluster network architecture, storage integration, and infrastructure design.
• Support the cluster bill of materials covering compute, networking, storage, and supporting infrastructure.
• Drive rack elevation, GPU server deployment, high-speed network topology, power readiness, and cooling readiness.

Contractor and field deployment

• Manage contractor onboarding, statements of work, scope definition, and delivery milestones.
• Oversee structured cabling, rack installation, equipment mounting, network and power connectivity, and hardware staging.

Validation and readiness

• Coordinate single-node GPU validation, multi-node cluster deployment, and P2P or RDMA interconnect validation.
• Drive benchmarking, stress testing, and performance verification before production release.
• Integrate monitoring and telemetry, prepare operational documentation and runbooks, and hand over clusters to operations teams.

Required qualifications

• 5+ years of experience in technical program management, infrastructure program management, or HPC infrastructure delivery.
• Experience with GPU cluster deployments or high-performance computing environments.
• Familiarity with GPU server architecture and distributed computing infrastructure.
• Experience defining system architecture and hardware bills of materials with Infrastructure Solution Architects.
• Experience managing data center hardware deployments and system integration.
• Ability to coordinate multi-vendor infrastructure projects across regions.

Preferred qualifications

• Experience deploying large-scale AI infrastructure or GPU clusters.
• Familiarity with InfiniBand, RoCE, or high-speed Ethernet networking and GPU interconnect validation.
• Experience with rack elevation, high-density rack deployment, cluster validation, or performance benchmarking.
• Background as a Systems Engineer, HPC Engineer, or Infrastructure Architect.

Additional advantages

• Experience deploying liquid-cooled GPU clusters or high-power racks.
• Experience with NVIDIA AI infrastructure platforms and AI training environments.

About UST