John Rankin

Altoona, WI
📞 (715) 296-0693
✉️ rankintoday@gmail.com

Summary

Software Development Manager and Senior HPC Engineer with 10+ years of experience building, automating, and operating large-scale high-performance computing systems. Deep expertise in HPC networking, cluster management, DevOps automation, and end-to-end system testing. Proven leader supporting mission-critical and exascale environments, driving quality, scalability, and operational efficiency.

Education

University of Wisconsin – Eau Claire
Bachelor of Science, Computer Science

Technical Skills

Programming and Automation: Python, Bash, Ansible,Java, JavaScript, Node.js, REST APIs

AI and HPC Infrastructure: GB200 and GB300 rackconfiguration, Azure, bare-metal and VM provisioning, configuration,automation, and administration

Networking and Interconnects: NVIDIA NVSwitch(interconnect), NVIDIA Spectrum-X (Ethernet), HPE Slingshot, InfiniBand, RoCE,Dragonfly and Fat Tree topologies, LAG/LACP, LLDP

Cluster Platforms and Systems: Slurm, Kubernetes, HPCM,Bright Cluster Manager, Linux (RHEL, Rocky, SLES), NFS, Lustre

Hardware Management and Provisioning: Redfish, IPMI,PXE, DHCP

Observability and Reliability: Prometheus, Grafana,telemetry pipelines, fabric counters, fault detection, anomaly detection

DevOps and Data: Git, Jenkins, Kusto/KQL, MySQL, NoSQL

Professional Experience

Microsoft — Remote

Senior Software Engineer | Azure AI Supercomputer Infrastructure
March 2026 – Present

• Provide operational platform support for Microsoft Fairwater, a purpose-built Azure AI datacenter architecture spanning Milwaukee and Atlanta and interconnecting 200,000+ NVIDIA GB200 GPUs as a unified supercomputer supporting OpenAI.
• Worked across Azure Core organizations to advocate for fleet operations that scale to massive datacenters, including O(1) power-cycle operations @ scale in the Azure control plane.
• Redesigned inefficient infra SW solution from scratch, covering cluster bring-up, health validation, AI workload readiness, distributed training support, and fleet-wide recovery.
• Executed large-scale AI workloads using Grok-1 to benchmark training-step intervals and identify infrastructure bottlenecks before handoff. Validated buildout readiness across NVLink, NVSwitch, NVIDIA Spectrum-X (MRC), and GPUs.
• Lead team analysis of AI training workload performance to identify metrics relevant to mean time between failures (MTBF), guide mitigations aimed at increasing mean time to failure (MTTF), and reduce mean time to repair (MTTR). Analyze ECC and FEC counters, power, thermal conditions, and GPU utilization using z-scores and Pearson and Spearman correlations where appropriate.
• Enabled root-cause analysis capabilities for large-scale GPU and HPC systems beyond XID issues, improving reliability, observability, and serviceability for production AI workloads.
• Helped resolve system issues as part of the team supporting TOP500 runs and record-breaking MLPerf Training benchmark results.

Hewlett Packard Enterprise (HPE) — Remote

Software Development Manager – HPC Networking
June 2022 – March 2026

• Own and lead development of all Switch CLI configuration utilities for the Slingshot networking product line.
• Own lab R&D, test, and development infrastructure supporting 30+ internal HPC systems.
• Partner with Product Management to plan initiatives, prioritize work, and align engineering resources with organizational goals.
• Lead DevOps and infrastructure automation efforts to provision bare-metal hardware efficiently, significantly reducing manual effort and turnaround time.
• Lead a team responsible for the Slingshot product testing suite, enabling continuous 24/7 end-to-end testing across:
  • Network bring-up and configuration
  • High availability
  • Dragonfly and Fat Tree topologies
  • Traffic validation, LAG/LACP, LLDP, and critical networking features
• Designed and drove development of an interactive Switch CLI integrated with API frameworks to improve usability and reduce R&D support burden.
• Own the R&D HPC system configuration tool used by internal stakeholders and customers, reducing cost and complexity for new system deployments and upgrades.
• Support exascale and marquee customers, leading incident response, root-cause analysis, and cross-team coordination.

Hewlett Packard Enterprise (HPE) — Remote

Senior Software Engineer – HPC Networking
August 2020 – June 2022

• Developed features for a microservices-based REST application, including packet forwarding from fabric managers to database services.
• Built internal tooling for system configuration, cable validation, fabric bring-up, and overall network health assessment.
• Maintained software deployments across standalone and Kubernetes environments, validating containers for each release.
• Created an internal documentation platform enabling collaborative Markdown contributions for Slingshot installation and troubleshooting guides used by customers.
• Designed and implemented a modular Slackbot integrating with Jenkins and internal systems to report build status and manage lab system reservations.
• Maintained Slingshot DNS infrastructure within Kubernetes environments.

Cray / Hewlett Packard Enterprise — Chippewa Falls, WI

Senior Software Engineer – HPC System Management, Configuration & Test
May 2014 – August 2020

• Designed and implemented end-to-end HPC cluster automation, from bare-metal installation through full software configuration using Python, Bash, Ansible, DHCP, and PXE.
• Reduced full system bring-up time from one week to approximately four hours through automation improvements.
• Defined installation and test processes for Bright Cluster Manager in large, distributed environments.
• Designed and implemented hardware testing suite software for HPC systems.
• Developed tools to analyze hardware performance and efficiency.
• Provided system configuration guidance and validation based on customer and internal requirements.
• Supported one-off hardware and software configurations under aggressive delivery timelines.
• Collaborated with Cray R&D on major platforms including Shasta (Kubernetes) and Urika-GX (OpenStack), aligning manufacturing requirements with engineering expectations.
• Troubleshot early hardware and software failures during new platform bring-up.
• Independently designed and built a custom HPC cluster management and testing solution capable of operating on 64+ cabinets / 1,000+ nodes, preventing shipment delays and avoiding significant financial penalties.

JB Systems LLC — Eau Claire, WI

Web Developer
September 2013 – May 2014

• Developed and maintained multiple web applications using the CodeIgniter PHP MVC framework.
• Designed and implemented MySQL database solutions.
• Delivered new features rapidly based on evolving client requirements.

Liberty Mutual Insurance — Wausau, WI

Software Developer Intern (SharePoint)
May 2013 – August 2013

• Built an internal web application from the ground up using SharePoint, C#, JavaScript, jQuery, and Knockout.js.
• Created deployment documentation and maintained internal Wikis.
• Developed a SharePoint 2013 App proof of concept.

Menards Inc. — Eau Claire, WI

Software Developer Intern
September 2012 – February 2013

• Contributed to a Spring MVC Java web application used in a commercial product.
• Implemented backend and frontend features in a production environment.