This workshop provides an overview of GPU programming history and architecture, comparing it with CPUs and discussing suitable applications. It introduces CUDA and its basics, emphasizing performance considerations and addressing common misconceptions. The abstract also covers implementing neural networks on GPUs, highlighting libraries like cuDNN and tools such as TensorRT for optimization. Additionally, it outlines an IoT course covering hardware aspects like ESP32 microcontroller usage and software components including MQTT setup and user interface design with JavaScript, CSS, and HTML.