Agriculture accounts for 70% of global freshwater consumption, making efficient irrigation systems essential in the face of increasing water scarcity. Traditional irrigation methods often result in significant inefficiencies, highlighting the need for sustainable alternatives. This study presents a hybrid smart irrigation system that integrates Internet of Things (IoT) sensors and machine learning (ML) algorithms to optimize water use without compromising crop yields. The system leverages real-time data from sensors measuring soil moisture, temperature, humidity, and wind speed, dynamically adjusting water distribution based on environmental conditions and crop requirements. Among the tested models, Decision Tree and XGBoost achieved the highest accuracy (98.92% and 98.88%, respectively) with minimal mean squared error. Results demonstrate up to 30% improvement in water-use efficiency and crop yield optimization compared to traditional methods. By addressing challenges such as high installation costs, data privacy concerns, and technological interoperability, this system offers a scalable and adaptable solution for sustainable precision agriculture worldwide.