DevOps Architect

Project Director - HPC / AI / GPU Infrastructure Deployment

Role Overview

We are seeking a seasoned Project Director – HPC / AI Infrastructure Deployment to lead large-scale, high-density compute programs involving GPU clusters, HPC workloads, and AI infrastructure. The role demands end-to-end ownership of deploying 10+ MW IT load data center environments, ensuring delivery of high-performance GPU-based compute platforms with cutting-edge networking and storage architectures.

Roles & Responsibilities

• Lead and deliver large-scale HPC / AI GPU cluster deployments (e.g., NVIDIA B200 / B300 GPU platforms) within defined timelines and budgets
• Drive execution of AI stack deployment (e.g., NVIDIA NVAIE) across hybrid/cloud/on-prem environments
• Manage multi-vendor ecosystems including OEMs, SI partners, and hyperscale technology providers
• Deploy and scale high-density GPU racks with liquid/air-cooled thermal strategies
• Design and oversee InfiniBand (IB) and high-speed Ethernet networks
• Experience with NVIDIA/Mellanox InfiniBand fabrics
• Configuration and optimization using UFM (Unified Fabric Manager)
• Strong understanding of BCM (Broadcom Ethernet switching) platforms
• Architect and implement Leaf-Spine network topology for ultra-low latency AI workloads
• Ensure effective integration of storage systems (parallel file systems, NVMe-based storage)
• Oversee deployment of Kubernetes-based GPU orchestration platforms
• Experience with containerized AI workloads and distributed training clusters
• Exposure to NVIDIA AI Enterprise (NVAIE), CUDA, and GPU virtualization frameworks
• Manage data center design, build, and repurposing for HPC workloads
• Oversee MEP (Mechanical, Electrical, Plumbing) systems implementation
• Enure optimized thermal management (liquid cooling, rear door heat exchangers, immersion cooling where applicable)
• Ensure optimized power density (kW/rack) planning
• Ensure optimized energy efficiency (PUE optimization)
• Establish robust governance frameworks aligned to:

a. HLD/LLD design validation

b. SOP adherence

c. Quality assurance benchmarks

• Implement risk mitigation strategies for large-scale deployments (supply chain, OEM dependencies, technology integration risks)
• Monitor program milestones and ensure SLA-based deliveries
• Drive structured cabling design (fiber-heavy HPC fabric, spine-leaf connectivity)

Qualifications & Experience

• B.E/B.Tech in Electrical / Electronics / Computer Science Engineering
• 15–25 years of experience in Data center infrastructure deployment, HPC / AI workload environments, large-scale IT infrastructure programs

Mandatory / Preferred Certifications

• PMP / PRINCE2 (mandatory for program governance)
• CDCP / CDCS / CDCPM certifications

Strongly preferred:

• NVIDIA AI Infrastructure / DGX / AI Factory certifications
• OEM certifications (Dell, HPE, Lenovo HPC systems)