In this paper, the researcher explores the use of Extreme Gradient Boosting (XGBoost) in weather forecasting with the aim of improving prediction accuracy, particularly for stormy weather conditions. By leveraging a comprehensive dataset comprising various meteorological variables within a structured data management framework, the research investigates the efficacy of integrating XGBoost into forecasting models. Through a series of meticulously designed experiments and evaluations, this study demonstrates the effectiveness of XGBoost in improving the accuracy of storm forecasts compared to traditional methods. The results indicate the enhancement in prediction sensitivity to indicate the highlight the weather patterns and predicting the storms. The findings of this research have a significant implication for the advancement of weather forecasting systems. Furthermore, the study highlights the importance of leveraging state-of-the-art machine learning techniques to address more complex forecasting challenges. The main role of Extreme Gradient Boosting in advancing the field of weather forecasting is to provide more valuable insights and opportunities for further development and research in this domain. The model's R-squared score of 0.62 and Root Mean Squared Error (RMSE) of 0.45 indicate a strong predictive performance when compared to baseline models. The F1-score, which measures the model's accuracy and recall balance, was 0.75, demonstrating how well it predicted storm occurrences. The study offers practical insight or improving forecasting functionality. These insights will inform the development of more reliable forecasting tools.