Ariful Azad

Ariful Azad serves as an Associate Professor in the Department of Computer Science & Engineering at Texas A&M University, where he leads research at the intersection of high-performance computing and graph analytics. His work focuses on developing scalable algorithms for graph machine learning with applications in bioinformatics and security informatics.

Educational Background:

• Ph.D. in Computer Science, Purdue University (2014)
• B.S. in Computer Science and Engineering, Bangladesh University of Engineering and Technology (2006)

Research Focus: Dr. Azad specializes in high-performance graph algorithms, particularly for distributed-memory systems. His pioneering work includes the Combinatorial BLAS library and novel approaches for graph neural networks (GNNs), with emphasis on explainability through Shapley values and optimization of sparse matrix operations. His bioinformatics research tackles large-scale metagenomics challenges through projects like Exabiome.

Publication Trends: Recent publications (2023-2025) reveal three dominant themes: (1) Scalable GNN explanation frameworks using distributed Shapley values, (2) High-performance sparse linear algebra for graph embeddings and knowledge graphs, and (3) Bioinformatics applications in metagenomics and network alignment. His work consistently bridges theoretical algorithm development with practical implementations for exascale systems.

Scientific Recognition:

• NSF CAREER Award (2024) for foundational contributions to scalable graph algorithms
• Indiana University Trustee's Teaching Award (2024)
• U.S. Department of Energy Early Career Award (2021)

Research Leadership: As principal investigator for multiple federal grants, Dr. Azad directs projects advancing graph analytics at extreme scales. His work on Weapons of Mass Destruction knowledge graphs demonstrates applied security research, while Exabiome represents significant contributions to computational biology. He actively develops open-source tools like PLANETALIGN for network analysis benchmarking, fostering reproducibility in computational science.