Mufrad Mahmud

My research focuses on developing Python-based computational models to analyze bulk-surface partitioning in atmospheric aerosols. Currently, as a Research Assistant at the Center for Atmospheric Research (ATMOS) at the University of Oulu, I am working on building scalable digital models that accurately represent the physical and chemical properties of acid surfactants. I specialize in machine learning, backend API development, and building automated data pipelines. My previous research includes integrating ML models and LLM-based insights for behavioral risk classification, as well as applying Convolutional Neural Networks (CNNs) to accurately detect and classify agricultural diseases. Through this work, I have developed a strong focus on computer vision and rigorous model validation, specifically ensuring that deep learning architectures accurately capture and interpret complex image data. I am also deeply interested in developing intelligent, data-driven platforms to solve complex problems in other domains, such as smart IoT systems and healthcare.