My research focuses on the design of intelligent, trustworthy, and scalable computational systems, spanning artificial intelligence, computer vision, language models, quantum computing, and hardware–software co-design. A central theme of my work is the integration of algorithmic intelligence with efficient and secure system architectures, ensuring that advanced AI techniques are both practically deployable and responsibly engineered.
In AI and computer vision, my research addresses representation learning, pattern recognition, and deep neural architectures for real-world perception tasks, including biomedical imaging, surveillance, and cyber-physical sensing. I actively explore data-efficient learning, robustness, and generalization, particularly in resource-constrained and noisy environments. Complementing this, my recent work engages with language models and multimodal AI, focusing on architectural efficiency, domain adaptation, and the alignment of language-centric intelligence with structured knowledge and decision-making systems.
A key pillar of my research is Explainable, Ethical, and Responsible AI (XAI & RAI). I develop methodologies for model interpretability, transparency, and bias mitigation, especially in safety-critical and socially sensitive applications. My goal is to move beyond post-hoc explanations toward design-level explainability, embedding accountability and trust directly into AI models and pipelines.
In parallel, I pursue research in quantum computing, emphasizing quantum algorithms, circuit synthesis, and quantum design automation, often leveraging machine learning for optimization. This work is closely aligned with my broader interests in VLSI CAD, FPGA-based systems, and embedded intelligence, enabling cross-layer innovation from algorithms to hardware.
Overall, my research vision is to bridge AI theory, systems engineering, and responsible deployment, contributing foundational methods and practical frameworks that enable next-generation intelligent systems to be powerful, efficient, interpretable, and socially trustworthy.
