This paper studies handwritten signature verification and the authenticity of a given signature. We have investigated the potential of ensemble model among other machine learning methods for application in the verification of signatures. The purpose of this study is to develop a web-based software that can authenticate the user’s signature, either genuine or forged. Comparison of the results achieved by our model with previously established models in the scope of our work on the basis of classification metrics, viz. accuracy, precision, recall, and F1-score. For the development of web application, Python, Django, TensorFlow, and React frameworks have been used. To get approved access, users will have to register themselves and log in using the web app features granted by the system. The validity of the signature is being determined by analyzing the user’s writing features, such as the length, width, and depth of each stroke. The ability to verify signatures more accurately over time will be enhanced using machine learning algorithms in the application.