Principal Architect – HPC & AI (NVidia Ecosystem)

Required Technical Expertise 

 

Core Mastery Areas 

• Expert level with deep architectural knowledge of NVIDIA data center platforms, including HGX and DGX platforms. 

• GPU-accelerated compute architecture for AI and HPC workloads. 

• High-performance networking architectures, especially with Spectrum-X. 

• Large-scale AI factory and HPC platform design. 

 

Storage Expertise 

• Hands-on architectural experience with high-performance parallel or scale-out storage systems. 

• Deep understanding of storage performance characteristics relevant to AI and HPC workloads, including bandwidth, IOPS, latency, and metadata scaling. 

• Proven experience integrating storage platforms such as VAST Data, Netapp, WEKA, DDN, or Lustre into GPU-accelerated environments. 

 

Working Proficiency 

• NVIDIA Base Command Manager (BCM) for cluster lifecycle management and operations. 

• Slurm for HPC workload scheduling and resource management. 

• Run:AI for GPU orchestration and multi-tenant AI workload optimization. 

• Kubernetes administration including deploying and managing GPU-accelerated AI and HPC workloads. 

• Linux systems administration in large-scale, performance-sensitive environments. 

• Containerized AI workflows and their interaction with schedulers and storage systems. 

 

Additional Experience 

• Experience optimizing existing HPC or AI platforms for performance, utilization, and cost efficiency. 

• Prior experience with multi-site, air-gapped, or regulated environments is beneficial but not required. 

• Experience with liquid cooling, power/cooling design, and data center integration strongly preferred. 

 

Leadership & Influence 

• Senior individual contributor role with influence through technical authority rather than people management. 

• Ability to mentor engineers and architects through design reviews, architectural guidance, and technical leadership. 

• Comfortable operating autonomously in complex, high-impact technical environments. 

 

Documentation & Repeatability Expectations 

• Develop and maintain high quality architectural documentation, including design blueprints, configuration guides, deployment validation reports, and operational runbooks. 

• Ensure all technical artifacts meet WWT’s One Voice standards for clarity, completeness, and technical accuracy, enabling consistent delivery across teams. 

• Create reusable templates, reference architectures, and standardized design patterns that accelerate future projects and improve delivery quality. 

• Drive a culture of documentation discipline, ensuring that every deployment is reproducible, supportable, and aligned with architectural intent.