This research technique focuses on dry bean classification using a machine-learning (ML) algorithm to enhance the quality of agricultural products, improve market value, and automate processes, thereby overcoming the limitations of conventional techniques. This technique supports the food industry business by improving the quality of agricultural products. In this work, seven types of dry beans are classified using a machine-learning algorithm via a web-based AI model. The machine learning algorithms were trained on 13,611 samples of the dataset. The dataset is processed through feature normalisation and outlier removal to improve ML model performance. The performance of the ML model was checked through the performance matrices, F1 score, accuracy and area under the curve (AUC). Through this performance metric, it is observed that CatBoost performs best to classify dry beans accurately compared to other ML models. In the experiment, different feature engineering and hyperparameter tuning refer to improving the classification score for dry bean types. The results showed that this AI web-based application accuratly classify dey beans, which help automization and quality check in the agricultural sector.