In this thesis, a highly robust and energy-efficient Wireless Multimedia Sensor Node (WMSN) is presented. This device was developed to be used in remote and distributed environments IoT applications. The major problem that has been addressed is that the transmission of multimedia information wirelessly consumes a lot of power. To solve this, a dual-radio The hardware platform is designed to be managed by an ESP32-S3 microcontroller (MCU). ESP32-S3 can be easily used to implement simple image processing algorithms and operate pre-trained neural network models on board itself. The device consists of 45 GPIOs, four SPIs, and three UARTs interfaces, which are modular and can be used to connect various peripheral sensors and devices.