This session discusses the deployment of AI systems on edge devices for real-time decision-making. Topics include model compression techniques, latency optimization, and low-power AI hardware (e.g., NVIDIA Jetson, Google Coral). Applications in smart cities, industrial IoT, and AR/VR will be explored. The session will also discuss federated learning and privacy-preserving computation. Participants will learn how to deploy efficient AI models to edge platforms using TensorFlow Lite, ONNX, and other tools, balancing performance with computational constraints.