Waste can be categorized majorly into dry waste and wet waste. The task of segregating these two is quite difficult and often requires human labor, which might involve handling hazardous materials. This research implements and evaluates an Automated Waste Segregation System, using advanced image classification techniques, such as YOLOv5 with IoT hardware to provide automated segregation between dry and wet waste. Our study focuses on training the YOLOv5s and YOLOv5l models on dataset with six different waste classes and evaluating for implementation. YOLOv5l demonstrated superior performance achieving mAP@ 0.5 of 57.6%, precision of 67.6%, and recall of 50.7% for all classes and precision of 93.3% and 87.7% for glass and metal classes, respectively. This system introduced a way to identify and separate wet from dry waste at the first step to optimize and enhance the waste management process.