My primary research interests lie in Explainable Artificial Intelligence and Natural Language Processing, focusing on Harmful Speech and Language Model Interpretability. I develop methods to enhance the intrinsic interpretability and reasoning capabilities of language models, ensuring faithful explanations and alignment with human values in high-stakes domains. I hold a Ph.D. and an M.Sc. in Natural Language Processing, both in Computer Science and Computational Mathematics from the University of São Paulo. I have an interdisciplinary academic background in computer science, linguistics, and computational mathematics. During my doctoral studies, I was a visiting researcher at the University of Southern California, and an invited speaker at the Leibniz Institute for the Social Sciences. My work has received national and international recognition, including the Google Latin America Research Award, the Maria Carolina Monard Award for the Best Thesis in Artificial Intelligence, the PROPOR Best Thesis Award in Natural Language Processing, and the Trevisan Prize for Students “AI for Good,” awarded by the Computer Science Department of Bocconi University and EDGE. I actively contribute to the international artificial intelligence and natural language processing research community, serving as a program committee member and area chair for venues such as ACL, AAAI, IJCNN, and ACM. I have also co-organized the International AAAI Conference on Web and Social Media, Workshop on Online Abuse and Harms, and Explainable Deep Neural Networks for Responsible AI, as well as shared tasks on hate speech detection [i] [ii]. With contributions to addressing hate speech and misinformation in underrepresented communities in the Global South, my research advances the interpretability and alignment of Large Language Models (LLMs) in real-world scenarios that promote accountability, fairness, inclusion, and the trustworthy deployment of NLP systems.