Eye tracking has been valuable for researchers in many domains. Most eye-tracking technologies have required an extra piece of costly hardware. Several other available eye-tracking solutions have usually not been very accurate and have required a costly subscription. We aimed our project at creating a free and open-source alternative that does not require specialized eye-tracking hardware and can be implemented using general camera modules. We developed a deep learning-based solution as a prototype for this project. Specifically, we developed a deep learning model to predict the gaze position of the user on the screen. We created our training dataset using a webcam to train the model. Each training sample consisted of a video frame of a person looking at the screen and the corresponding true gaze position on the screen. By using a model with a gaze location as output on a 2D plane, the deep learning model became capable of interpreting and predicting gaze positions in most cases. Our results have demonstrated the computational power of the deep learning algorithm that can be used effectively to create and deploy eye-tracking solutions.