Principal HPC Architect

About the role

The Principal HPC Architect designs, builds, optimizes, and supports large scale compute environments used for scientific computing, AI/ML workloads, simulation, and data intensive research. This role blends systems engineering, performance tuning, cluster architecture, and hands-on troubleshooting. The engineer partners with researchers, developers, and IT teams to deliver reliable, scalable, and high performance compute infrastructure.

Key Responsibilities

• HPC Architecture & Engineering
  • Design and implement HPC clusters, including compute, storage, networking, and job-scheduling components.
  • Evaluate and integrate new technologies (GPUs, accelerators, interconnects, filesystems).
  • Develop automation for cluster provisioning, configuration, and lifecycle management.
  • Architect solutions for large-scale parallel workloads, AI/ML pipelines, and data-intensive applications.
• Performance Optimization
  • Profile and tune applications for CPU, GPU, memory, and I/O performance.
  • Optimize MPI, OpenMP, CUDA, and other parallel programming frameworks.
  • Benchmark hardware and software stacks to guide procurement and architecture decisions.
• Operations & Reliability
  • Maintain and monitor HPC clusters, job schedulers (Slurm, PBS, LSF), and distributed filesystems (Lustre, GPFS, BeeGFS).
  • Troubleshoot complex system issues across compute, storage, and network layers.
  • Implement security best practices, patching, and compliance controls.
  • Ensure high availability and efficient resource utilization.
• Automation & DevOps
  • Build and maintain CI/CD pipelines for HPC-related software and infrastructure.
  • Use tools such as Ansible, Terraform, Kubernetes, or custom scripts to automate workflows.
  • Develop monitoring and observability solutions (Prometheus, Grafana, ELK, etc.).
• Collaboration & Leadership
  • Work closely with researchers, data scientists, and engineering teams to support workload optimization.
  • Provide technical leadership, mentorship, and guidance to junior engineers.
  • Document architectures, procedures, and best practices.
  • Participate in capacity planning and long-term HPC strategy.

Required Qualifications

• Extensive experience with Linux systems engineering in large-scale compute environments.
• Solid understanding of distributed systems and cloud infrastructure.
• Deep knowledge of HPC schedulers (Slurm preferred), MPI stacks, and parallel computing models.
• Strong understanding of high-speed interconnects (InfiniBand, RoCE) and distributed storage systems.
• Proficiency in scripting languages (Python, Go, Bash) and automation frameworks.
• Experience with GPUs (NVIDIA CUDA, MIG, NVLink) and accelerator-based computing.
• Familiarity with containerization (Singularity/Apptainer, Docker) in HPC contexts.
• Strong troubleshooting skills across hardware, OS, and application layers.
• Understanding of networking fundamentals (TCP/IP, DNS, load balancing).
• Background in high-availability and distributed systems at scale.

Soft Skills

• Excellent communication and cross-functional collaboration.
• Ability to translate research needs into technical solutions.
• Strong ownership mindset and ability to lead complex initiatives.

Base Pay Range

$162,700.00 - $284,700.00 Annually

Benefits

• Medical, dental, vision, life, and other voluntary benefits.
• 401(K) including company matching.
• Employee stock purchase program (ESPP).
• Student debt assistance.
• Tuition reimbursement program.
• Development and career growth opportunities and programs.
• Financial planning benefits.
• Wellness benefits including an employee assistance program (EAP).
• Paid time off and paid company holidays.
• Family care and bonding leave.