Accurate rainfall forecasting is critical for mitigating the risks of floods, landslides, and ensuring efficient agricultural and water management, especially in regions like Kerala, where monsoons significantly impact the environment and economy. Traditional statistical models struggle to capture the complex, nonlinear nature of atmospheric processes. To evaluate the model's performance, we compare the predicted rainfall with the actual recorded data using statistical metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and accuracy. This project aims to improve rainfall forecasts by using machine learning techniques, specifically Long Short- Term Memory (LSTM), to create a predictive model. A dataset consisting of Kerala state meteorological data from 2010 to 2023 is used for training the model. Meteorological data from Kerala (2010–2023), including temperature, humidity, wind speed, and historical rainfall, was collected from the Shiny Weather website and preprocessed for training the Machine Learning model. The application of Long Short-Term Memory (LSTM) allows for better handling of large datasets and more accurate predictions by accounting for the complex interactions between weather variables. By providing real-time rainfall forecasts, this tool will contribute to better preparation for extreme weather events and aid in making informed decisions across agriculture, water management, and disaster preparedness domains.