This document describes a research project on controlling a computer using hand gestures. The researchers created a real-time gesture recognition system using convolutional neural networks (CNNs). They developed a dataset of 3000 training images of 10 different hand gestures for tasks like opening apps. A CNN model was trained to detect hands in images and recognize gestures. The model achieved 80.4% validation accuracy and was able to successfully perform operations like opening WhatsApp, PowerPoint and other apps based on detected gestures in real-time. The system provides a cost-effective and contactless way of interacting with computers using hand gestures only.