Raphael Mazzine
Building what matters with AI
Ph.D. in Applied Economics with 14+ years turning research into intelligent systems that solve real problems.
About
Researcher and engineer with a multidisciplinary background spanning nanotechnology, physical chemistry, and applied economics. I build AI-driven products at the intersection of science and software — from explainable ML to large-scale language model applications.
Career
Elsevier
Data Scientist
Developing AI solutions for educational content, applying methods to enhance trustworthiness and efficiency in one of the world's largest scientific and educational publishers.
Complexio
Data Scientist
Optimized AI models and built software solutions to automate business processes, delivering efficient machine learning pipelines and intelligent automation for clients across multiple industries.
UL Solutions
Data Scientist
Applied machine learning and data analytics to safety science, developing predictive models and data-driven insights for product certification and compliance.
University of Antwerp
Doctoral Researcher
Conducted Ph.D. research in Explainable AI and Counterfactual Explanations at the Faculty of Applied Economics, publishing peer-reviewed work on making ML models interpretable.
IBM
Data Scientist & AI Engineer
Designed and deployed machine learning solutions for enterprise clients, working on NLP pipelines, predictive analytics, and AI-powered automation at scale.
Education
Applied Economics — Artificial Intelligence
University of Antwerp
Physical Chemistry
UNESP — São Paulo State University
Nanotechnology
UFRJ — Federal University of Rio de Janeiro
Academic Publications
Peer-reviewed research in Explainable AI, Counterfactual Explanations, and Applied Machine Learning
A model-agnostic and data-independent tabu search algorithm to generate counterfactuals for tabular, image, and text data
European Journal of Operational Research
DOI: 10.1016/j.ejor.2023.08.031Explainable image classification with evidence counterfactual
Pattern Analysis and Applications
DOI: 10.1007/s10044-021-01055-yA Framework and Benchmarking Study for Counterfactual Generating Methods on Tabular Data
Applied Sciences
DOI: 10.3390/app11167274Pseudocapacitance phenomena and applications in biosensing devices
Electrochimica Acta
DOI: 10.1016/j.electacta.2019.03.083Latest Articles
Insights on applied AI, RAG systems, LLM efficiency, and AI in education
