Rapid urban growth has intensified environmental challenges in major Indian cities, especially the Urban Heat Island (UHI) effect. Pune and its surrounding urban regions have experienced a noticeable increase in land surface temperature because of uncontrolled urban expansion, decline in green cover, and increased built-up surfaces. This research presents an AI-enabled framework integrating Remote Sensing, Geographic Information Systems (GIS), satellite thermal observations, and Machine Learning techniques for identifying and analysing Urban Heat Island patterns in and around Pune city. Thermal and multispectral datasets obtained from Landsat 8/9, MODIS, and Sentinel-2 satellites were used to estimate Land Surface Temperature (LST), vegetation conditions, built-up density, and land-use characteristics. Machine learning approaches such as Random Forest, Support Vector Machine (SVM), and clustering methods were employed for hotspot classification and prediction of future heat-vulnerable zones. The findings reveal that heavily urbanized locations show substantially higher temperatures compared to vegetated and water-covered regions. The developed AI-GIS model can support urban planners, environmental agencies, and municipal authorities in implementing sustainable climate adaptation and heat mitigation strategies.