Accurate weather forecasting is a critical challenge influenced by the dynamic and multifaceted nature of atmospheric conditions. This study introduces a hybrid weather prediction model that utilizes machine learning (ML) and deep learning (DL) techniques to enhance accuracy and reliability. The proposed system integrates Random Forest, Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) networks to address the limitations of traditional weather forecasting methods and standalone ML/DL models. Random Forest provides robust predictions for tabular weather data, CNNs analyze spatial patterns in satellite imagery, and LSTMs capture temporal trends in sequential data. This ensemble approach ensures comprehensive and reliable weather predictions. Advanced preprocessing techniques, including noise reduction, normalization, and outlier detection, enhance the quality of input datasets. The system also prioritizes scalability and energy efficiency through distributed computing and model optimization techniques like pruning and quantization. The hybrid model demonstrates superior performance in predicting both average and extreme weather conditions, such as heatwaves and storms, across diverse geographic regions and climatic conditions. A user-friendly interface provides real-time forecasts and customizable alerts accessible via web and mobile platforms. Practical applications span agriculture, transportation, disaster management, and more, enabling informed decision-making and improved preparedness. This research highlights the significance of integrating advanced predictive models with user-focused design to create scalable, efficient, and actionable weather forecasting solutions. Future directions include incorporating additional data sources, enhancing edge computing capabilities, and expanding support for diverse platforms to meet evolving meteorological challenges.