I will present an overview of my research experiences with Machine Learning (ML) in the areas of Hydrology, Natural Disasters, and Geosciences. We will talk about the main problems that are encountered when modeling environmental systems with ML and how to avoid or solve them. We will also briefly talk about some popular ML methods and their particularities, and innovations in the area of Explainable Artificial Intelligence. Afterwards, we will have a moment for questions, comments, and discussions. PT-BR only.