Summary
Paul Nowoczynski is a seasoned storage systems architect with two decades of experience building high-performance distributed storage for HPC, AI/ML, and enterprise environments. He led development of breakthrough systems like DDN’s IME—the first to exceed 1 TB/s—and earlier pioneered burst-buffer concepts (Zest) that influenced widely cited research, demonstrating a rare ability to turn research ideas into production-scale deployments. Paul combines deep expertise in erasure coding, parallel filesystems, and topology-aware data placement to deliver extreme throughput, low latency, and efficient TCO across large clusters. Currently he is designing a clustered block storage platform that marries novel erasure codes with node-local and top-of-rack placement to accelerate training, inference, and real-time scientific workflows. His career blends hands-on systems engineering at places like PSC and DigitalOcean with entrepreneurial freelance work (e.g., PumiceDB), making him equally fluent in prototyping novel algorithms and shipping robust production systems.
10 years of coding experience
11 years of employment as a software developer
BS, Information Science, BS, Information Science at University of Pittsburgh