Complex Event Processing (CEP) deals with collecting events from multiple sources, detecting patterns, filtering, transforming, correlating and aggregating them into complex events. Predictive analytics (PA) on the other hand deals with anticipating (forecast) the occurrence of events so to be able to react before these occur. We developed a system which effectively predicts thunderstorms by ingesting real-time streaming data and by using CEP and PA.  For this we used weather data from National Oceanic and Atmospheric Administration (NOAA) and perform Machine Learning and other complex events to detect the patters as they happen.