This book chapter will provide the transformative role of time series modeling in Machine Learning (ML) approach in enhancing weather forecasting accuracy in India. The climate changing patterns have more complexity information, that’s why, traditional forecasting techniques sometimes fells to capture the non-linear dynamics of such cases. The chapter presents an overview of key ML models—including Artificial Neural Networks (ANN), Random Forest (RF), Long Short-Term Memory (LSTM), and Extreme Gradient Boosting (XGBoost) that have been widely adopted for forecasting different meteorological variables. Also, we present the methodologies architecture and mathematical expression of each ML models. ANN and LSTM are highlighted for their effectiveness in learning temporal dependencies in time series data, while RF and XGBoost are recognized for their high predictive accuracy and robustness with decision trees. The chapter also discusses model evaluation techniques, optimum hyperparameter selection and general python codes for real-time data prediction in the context of weather systems. This chapter underscores the potential contributing to better decision-making and sustainable development across weather-sensitive sectors.