Senior HPC Architect

About the role

As a Senior HPC Architect at Oak Ridge National Laboratory (ORNL), you will lead the architecture, design, and evolution of high-performance computing platforms supporting both open research and classified computing missions.

Responsibilities

• HPC System Design & Architecture
  • Lead the design and deployment of HPC systems to meet performance, reliability, and security requirements for research and/or classified computing environments.
  • Define and maintain reference architectures for compute, network, storage, and platform services, including lifecycle planning and roadmap development.
  • Produce and maintain technical documentation: architecture diagrams, configuration standards, operational procedures, and engineering decision records.
  • Guide system and accelerator architecture decisions with a strong grasp of how GPU/accelerator architecture impacts datacenter and AI/HPC workloads (including LLM-adjacent workloads).
• Automation, Infrastructure Platforms & Enablement
  • Identify automation targets and lead adoption of infrastructure-as-code and configuration management (e.g., Ansible, Puppet, Chef, Kickstart, Satellite).
  • Build standardized, repeatable deployment workflows; contribute to internal platform tools and codebases where appropriate (e.g., Python-based automation).
  • Where applicable, architect and support container and platform capabilities (e.g., Docker/Kubernetes) to enable reproducible scientific workflows and service deployment.
• Leadership & Collaboration
  • Lead HPC-related projects from planning through implementation and steady-state operations; manage technical risks, dependencies, and milestones.
  • Partner with scientists, researchers, and mission stakeholders to translate workflow requirements into platform capabilities.
  • Mentor junior engineers; create knowledge-sharing practices, documentation, and engineering standards.

Qualifications

• BS degree in Computer Science, Engineering, or a related field and a minimum of 8+ years of relevant experience (or equivalent combination of education and experience).
• 8+ years of experience in HPC engineering with demonstrated strength in system architecture, cluster operations, parallel computing environments, and performance optimization.
• Demonstrated experience working in high-security and/or regulated environments.
• Strong experience with HPC cluster management and scheduling.
• Experience with HPC performance monitoring and benchmarking using tools such as Grafana, Nagios, Ganglia (or equivalent).
• Ability to lead technical initiatives, write clear technical documentation, and communicate effectively with both engineering and non-engineering stakeholders.

Preferred Qualifications

• Familiarity with parallel file systems / advanced storage.
• Experience with containerization and HPC-adjacent platforms (e.g., Docker, Kubernetes, Kubeflow) in a way that complements scheduler-based HPC usage.
• Experience with virtualization platforms (e.g., VMware) in support of HPC infrastructure services.
• Strong Infrastructure-as-Code background (e.g., Ansible, plus cloud/IaC such as Terraform/Packer where relevant).
• Experience supporting scientific software development and deployment and research user workflows.
• Preferable: experience with geospatial data workflows, including large geospatial/raster/vector datasets, spatial ETL pipelines, and performance considerations for geospatial analytics at scale (e.g., tiling/partitioning strategies, I/O patterns, reproducibility, and access controls for sensitive geospatial data).
• Strong leadership, mentoring, and cross-team coordination skills; ability to manage multiple priorities in fast-paced, high-consequence environments.