Amin Shafiee

Papers

Total Citations

H-Index

About

Amin Shafiee is a rising researcher at the forefront of next-generation computing architectures, specializing in the intersection of graph neural networks (GNNs) and silicon photonics. His most notable contribution is the development of **GHOST**, a pioneering accelerator that leverages silicon photonics to dramatically speed up GNN computations—a critical need for applications ranging from recommendation systems and social network analysis to drug discovery and robotics. By harnessing light instead of electrons, GHOST addresses the severe memory and communication bottlenecks that plague traditional electronic accelerators when processing large-scale, irregular graph data. This work has already garnered significant attention (over 12 citations in its first year alone), marking Shafiee as a key innovator in the push toward ultra-efficient, domain-specific hardware. His research bridges the gap between photonic device physics and practical machine learning systems, offering a glimpse into a future where complex graph analytics can be performed with unprecedented speed and energy efficiency. For students and researchers, Shafiee’s work represents a compelling case study in how cross-disciplinary thinking—combining photonics, computer architecture, and AI—can solve grand challenges in computing.