In many areas in rural, disaster-prone and tourist of Bangladesh, clean and safe drinking water supply is seriously hampered — specially during the time of floods, power outages and disruption of communication facilities. The above investigative study has discussed an economically sophisticated, efficient and smart solar driven water quality monitoring system using IoT (LoRa) Technology. Several sensors and an ESP32 microcontroller are used to collect sensor data from the environment to cloud platforms. Field testing verified long-range data delivery at RSSI levels of –85 dBm or greater over 200 meters, even in harsh environments. Six machine learning models were assessed with water quality classification — Decision Tree, Random Forest, Naive Bayes, Logistic Regression, K-Nearest Neighbors and SVM. The best performance with Decision Tree and Random Forest was in obtaining a 99.1% accuracy, which was also able to achieve precision, recall, and F1 scores of above 0.99. An easy-to-use web-based dashboard was also developed to display live sensor readings, current solar status, and machine learning (ML) water quality predictions, and to send real-time alerts when contamination is detected. This approach was tested in field trials conducted in remote offline areas, which validated that this system is universal, easily scalable and can be implemented on top of existing public health networks to help the environment and be economically viable.