The project focuses on predicting coffee exports in Colombia through a comprehensive approach involving advanced data manipulation, the application of machine learning models, and the construction of an interactive dashboard in Power BI. The complexity of the data transformation process was addressed using tools such as Python and the Pandas library, optimizing and reducing data to manage computational resource limitations. Careful interrelation of climate and export data overcame initial technological challenges, highlighting the importance of appropriate infrastructure. The selection and adjustment of the BaggingRegressor model were conducted through exhaustive evaluation based on key metrics such as R-Squared and RMSE.