Doost-Mohammady’ research lies at the intersection of wireless networking, machine learning, and reconfigurable computing, emphasizing software-defined systems. His current work focuses on developing intelligent, high-efficiency wireless networks that utilize AI-driven resource allocation, RAN virtualization, and large-scale MIMO architectures. He has made significant contributions to various research platforms, including RENEW, 3DML and more recently ETHOS  and Houdini. In these projects, he designs and implements end-to-end systems that integrate real-time machine learning algorithms with advanced wireless protocols, aiming to optimize performance in next-generation networks, including 5G and beyond.