<p>Advanced knowledge of weather systems is essential for various routine activities, as sudden precipitation events can disrupt social life and affect multiple sectors. Weather radar, which provides detailed spatial and temporal variations of cloud systems, is a crucial tool for weather prediction models. Radar reflectivity data from radar can be converted into precipitation rates, offering prior information useful for hazard mitigation. This capability is vital for mitigating hazards associated with severe weather events such as floods, thunderstorms, and hailstorms. A reliable nowcasting system that captures the nonlinear and rapidly evolving nature of weather systems can significantly reduce damages and losses caused by severe weather. In this study, we present an indigenously developed radar-based nowcasting system utilizing the python framework for short term ensemble prediction (PySTEPS), applied over the Trivandrum region in India. This model generates nowcasts from radar reflectivity data obtained from the Thumba Equatorial Rocket Launching Station’s (TERLS) C-band radar, which is preprocessed using a tailored clutter-filtering approach to mitigate site specific artifacts. It employs optical flow techniques and cascade decomposition to accurately depict the spatial and temporal evolution of cloud systems. To address uncertainties in precipitation forecasting, the system generates 15-minutes ensemble forecasts for a lead time of up to 90&#xa0;min. To determine the accuracy of our nowcasts, we validated three months of nowcast data (June, July, and August) against observations. Validation metrics included the confusion matrix and skill scores such as Fractional Skill Score (FSS), Heidke Skill Score (HSS), Probability of Detection (POD), False Alarm Ratio (FAR), and Critical Success Index (CSI). Additionally, we present case studies demonstrating the model’s operational nowcasting capabilities. The results indicate that our model shows promising potential for operational precipitation nowcasting.</p>