Hand gesture recognition is an important research area with applications in various fields, such as human-computer interaction, virtual reality, and robotics. In this paper, we propose a hand gesture recognition system using the transfer learning technique of AlexNet. The proposed system uses a pre-trained deep convolutional neural network, AlexNet, which is fine-tuned to recognize hand gestures. The proposed system is evaluated on a hand gesture dataset, which consists of 3 different hand gestures representing the Door Lock, Lights, and Security System, and we specially add a neutral class to remove ambiguity in results. The experimental results are tested in real-time and show good performance. The performance of the proposed system is compared with other state-of-the-art deep learning- based hand gesture recognition systems, and the results show that the proposed system proves to be better. The application of the system includes an Automated Home system, a Security System, etc.