Face detection is a fundamental problem in computer vision with applications in security, authentication, and human-computer interaction. This thesis presents a prototype of a real-time face detection system specifically designed for detecting a driver’s face inside a vehicle cabin using the Multi-task Cascaded Convolutional Network (MTCNN) deployed on an ESP32-S3 microcontroller. The objective of this work was to design and implement an embedded system capable of performing accurate and efficient face detection and landmark localization under resource-constrained conditions. A custom-trained MTCNN model was developed using newly integrated tools to enable deployment on TensorFlow Lite for Microcontrollers (TFLM). Experimental results demonstrate that the optimized model achieves reliable detection performance with acceptable latency and power consumption on the ESP32-S3 microcontroller.