Summary
Marty Kandes is a computational scientist and Senior Research Specialist with 10+ years designing and optimizing high-performance computing solutions for scientific workloads at the San Diego Supercomputer Center. He blends deep domain expertise in computational atomic physics, quantum mechanics, and numerical PDEs with practical experience in distributed systems, HPC infrastructure, and multi-node GPU benchmarking. Marty has driven large-scale sustained-performance benchmarks on leadership-class systems and led I/O and multi-GPU performance studies comparing BeeGFS, Lustre, and WekaIO, bringing a measurement-driven approach to system tuning. He also has operational pedigree running global high-throughput middleware (GlideinWMS/HTCondor) and has exposed subtle system vulnerabilities and resiliency issues through careful testing. Comfortable from low-level numerical algorithm design to datacenter operations and cloud integration, he pairs a PhD-level research background with hands-on engineering and mentoring experience. An often overlooked strength is his track record of translating physics experiment instrumentation and control work into robust software and modeling tools for HPC and distributed computing.
10 years of coding experience
9 years of employment as a software developer
Bachelor of Science - BS, Applied Mathematics, Bachelor of Science - BS, Applied Mathematics at University of Michigan
Doctor of Philosophy - PhD, Computational Science, Doctor of Philosophy - PhD, Computational Science at San Diego State University