There are many examination portals available in the world today. These portals are spread across multiple servers and are used to administer online tests for a variety of purposes. Some of these purposes include entrance exams, national and international competitions, and placement tests. However, from what we've observed, nearly every site is made to administer exams with multiple choice questions. Here, developing extremely rare technology is our goal rather than working on already-existing technology. We are discussing the online descriptive examination system here. Because multiple choice questions consist of a question, a few options, and a field that saves the correct option in the database all inside the same question, they are simple to answer. However, this is untrue for descriptive queries. In order to allocate marks to responses, it incorporates or makes use of the ideas of natural language processing, or NLP. Answers are just strings, and the model's task is to manipulate the answer string in order to assign the proper marks to the examinee's answers. In essence, the information is gathered using an online descriptive examination method. It is then examined further, and the developed model gives the questions' responses precise grades. The back-end is implemented in Python using the Django web framework, the NLTK natural language processing package, and SQLite version 3 for database purposes. The front-end uses HTML version 5, CSS version 3, Bootstrap, and Javascript.