Weather and climate conditions have become more erratic and unpredictable, both domestically and internationally, which might have a devastating effect on agricultural output. Numerous climatic elements, including temperature, humidity, precipitation, air quality, and many more, are constantly altering in a frighteningly unpredictable way. Having a local weather station that can provide farmers with up-to-date information on the weather is crucial. It is essential to have a nearby, real-time weather station that can inform farmers of the present weather. In order to overcome the se issues, a novel Adaptive Weather forecasting Application using REal time sensors for AGRIculture (AWARE-AGRI) technique has been proposed in this work. The proposed AWARE-AGRI technique monitors the weather in real time by using Stacked convolutional neural network (SCNN) based Bidirectional Long Short-Term Memory (BiLSTM) technique for classifying weather data. An Android application has been deployed for tracking the weather online, which can access a dedicated server. MATLAB has been used to evaluate the proposed technique. Certain metrics, including accuracy, precision, f1-score, and recall have been used to evaluate the performance technique. The proposed AWARE-AGRI will give farmers more hope that they will be able to complete their agricultural responsibilities in real time.