This thesis presents the design and implementation of a machine learning (ML)-based bathroom humidity control system aimed at energy efficiency and preventing mold growth with causes structural damage. Conventional bathroom ventilation systems rely on manual controls or fixed humidity thresholds, which often result in excessive energy use or insufficient ventilation. The developed system integrates an ESP32 microcontroller with temperature, water flow, and humidity sensors to monitor environmental conditions in real time. A lightweight ML model, trained using TinyML techniques, is deployed on the ESP32 to forecast short-term humidity levels based on live sensor data. Using the predicted humidity trends, the system proactively activates an exhaust fan via a relay module before humidity exceeds critical levels, thereby ensuring timely ventilation.Experimental evaluation demonstrates that the system effectively predicts humidity surges and responds with optimized fan control, offering a cost-efficient, scalable, and privacy-preserving solution for smart home ventilation.