This presentation explores the advancement of rainfall analysis through the integration of artificial intelligence (AI), remote sensing, and IoT-based observation, with applications spanning from urban flood risk reduction to regional hydrological resilience. Building on high-resolution precipitation data and sensor networks, this work demonstrates how AI (e.g., deep learning models, image fusion techniques, etc.) improves the accuracy of rainfall estimation over conventional approaches. In addition, the presentation highlights recent developments in rainfall frequency analysis using short-duration data, where bivariate and nonstationary models have been employed to enhance the reliability of design rainfall estimation under changing climatic and urban conditions. By combining statistical frequency approaches with spatiotemporal monitoring from satellite and IoT sensors, this research establishes a multi-scale framework for early warning, flood mapping, and infrastructure design. The integrated methodology not only supports real-time flood forecasting in urban environments but also contributes to long-term regional planning for climate adaptation and water system resilience.