In the realm of modern agriculture, optimizing crop yield and fertilizer management is crucial for enhancing productivity, sustainability, and resource utilization. This study explores the application of the Random Forest mechanism, a machine learning model identified for its versatility as well as robustness, to predict crop yields and optimize fertilizer usage in various agricultural settings. By analyzing extensive datasets comprising agronomic variables such as soil properties, crop types, and historical yield records, the Random Forest algorithm constructs predictive models using an ensemble of decision trees. Experimental evaluations on real-world agricultural datasets demonstrate the superiority of Random Forest over conventional approaches in yield prediction accuracy and fertilizer optimization efficiency. The findings highlight the potential of machine learning techniques to revolutionize agricultural practices, offering data-driven insights for decision-making and promoting sustainable crop management across diverse cropping systems.