Saman Biookaghazadeh is a research engineer at Google DeepMind with 11 years of experience designing high-performance distributed systems for AI at scale. He helped build Google Cloud’s first A3+ H100 GPU cluster and authored RDMA enhancements for NCCL to enable GPU-to-GPU communication across thousands of accelerators. Saman also developed a Google-specific GPUDirect pipeline for direct storage-to-GPU data loading now used by DeepMind and led a storage optimization project whose weighted-chunk allocation algorithm saved Google millions by balancing I/O across heterogeneous drives. With a Ph.D. in High Performance Computing, his research focuses on scheduling across CPUs, GPUs, and FPGAs to squeeze more throughput from heterogeneous clusters. He has driven both product-grade system integrations and low-level infrastructure innovations, and has hands-on experience compressing and optimizing models for edge ARM devices. Based in New York, he blends deep academic rigor with production engineering that directly accelerates large-scale AI training.
11 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Systems, Doctor of Philosophy (PhD) Computer Systems at Arizona State University
Bs computer engineering, Bs computer engineering at University of Tehran
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Florida International University
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.
Request Free Trial
Saman Biookaghazadeh - Research Engineer at Google DeepMind