Abbas (Yasin) Mazloumi

About

Yasin Mazloumi has a broad background in Computer Architecture, GPU Architecture, and Distributed Graph AI/Analytics, having dedicated his career to advancing the field of high-performance computing (HPC). Prior to joining the Kempner Institute, his research at the University of California, Riverside, focused on Distributed Graph Analytics, where he developed distributed and scalable high-performance solutions for graph processing by employing resources available on heterogeneous computing clusters. His expertise extends to leveraging the new parallel programming paradigm of running simultaneous graph queries to amortize the distributed computing cost over a batch of queries, effectively tackling complex computational challenges. These experiences have equipped him with the skills to innovate and optimize high-performance computing systems, making significant strides in developing efficient and effective platforms.

Research Focus

Mazloumi is a Senior ML Research Engineer in the Kempner Institute’s Research and Engineering team. He focuses on building and optimizing large AI models that can understand language, answer questions, and learn from complex environments with specialization in distributed training and parallel computing, designing systems that allow these models to train efficiently at scale on state-of-the-art AI clusters. Additionally, he explores how principles of natural intelligence and neuroscience can inspire new AI computing paradigms across both software and hardware. At the Kempner Institute, he also contributes to collaborative teaching and research by co-developing workshops that make advanced research and engineering more accessible and impactful to support the broader AI community.