Summary
Mario Dudjak is an assistant professor and AI/ML lead based in Osijek with nine years of experience bridging academic research and industry-grade machine learning solutions. His research expertise centers on supervised classification and learning from class-imbalanced data, complemented by a niche focus on integrating bio-inspired optimisation (evolutionary and swarm methods) into feature selection, model tuning, and preprocessing. He has moved steadily from PhD researcher to postdoc and faculty roles while teaching and supervising practical data science projects, and he applies that research directly in industry work such as credit-risk modelling. An active reviewer for international journals, Mario combines rigorous publishing credentials with hands-on software engineering experience from web development to production ML systems. He brings a practical, experiment-driven mindset to improving ML pipelines—and often looks to nature-inspired algorithms for solutions that standard approaches overlook.
9 years of coding experience
4 years of employment as a software developer
Prirodoslovno-matematička gimnazija, Prirodoslovno-matematička gimnazija at Srednja škola Isidora Kršnjavoga Našice
Doctor of Philosophy - PhD, Artificial Intelligence, Doctor of Philosophy - PhD, Artificial Intelligence at Josip Juraj Strossmayer University of Osijek
English, Croatian, Slovak, German