Rafael Lopes Almeida

I studied Electrical Engineering as an undergraduate and got drawn into machine learning along the way. After graduating, I spent a few years in industry, working as a Computer Vision Engineer, Data Scientist, and AI Engineer. I worked on license plate recognition, fraud detection, NLP systems, and predictive toxicology.

For my Master’s degree I built machine learning models for QSAR (Quantitative Structure-Activity Relationship) applications in pharmaceutical research. That work pulled me into cheminformatics and, more broadly, into thinking about how AI methods can be useful in drug discovery. I have also been involved as a P&D Researcher in industry and as a Research Collaborator at the Computational Intelligence Laboratory (LITC), where the focus has been on developing machine learning approaches with practical applications in cheminformatics.

At UEF, my research sits at the intersection of AI and drug discovery. I work with graph-based molecular representations and graph neural networks (GNNs), and I’m particularly interested in model explainability, that is, understanding why a model makes a given prediction, not just whether it gets the right answer. I also develop machine learning and deep learning methods aimed at improving predictive models for pharmaceutical research, with a focus on data-intensive problems in drug discovery pipelines.