The worldwide pandemic significantly disrupted education systems, especially for engineering students from Bangladesh. In this pandemic scenario, it is already challenging for an institution to predict a student's success. It has never been easy to predict student performance, and the pandemic has made it far more difficult. The primary goal of this work is to use a typical machine learning technique to develop a machine learning model that predicts the way the pandemic would impact students' performance. Information is collected from students at a few private universities in Bangladesh once a few appropriate features have been selected. This study examines ten different parameters, such as total credit for the semester, credit for theory and lab subjects, CGPA, SGPA, financial difficulties, etc. The suggested model is tested and trained using the preprocessed data. We test the model extensively using seven different machine learning techniques. Lastly, the experiment's complete results are displayed, validating the machine learning model's effectiveness in this kind of application. The Decision Tree Regressor outperforms the other regression models with an accuracy of 95.57%.