The agricultural sector is highly vulnerable to the adverse impacts of climate change and the excessive use of pesticides, posing a serious threat to global food security. Accurate crop yield prediction is crucial for mitigating these risks and promoting sustainable agricultural practices. This research proposes a novel crop yield prediction system that integrates one year’s worth of meteorological data, pesticide usage records, and crop yield statistics using machine learning techniques. Comprehensive data preprocessing was performed, including data collection, cleaning, and enhancement, followed by the training and evaluation of three machine learning mod- els: Gradient Boosting, K-Nearest Neighbors, and Multivariate Logistic Regression. To optimize model performance and prevent overfitting, GridSearchCV was used for hyperparameter tuning across K-Fold cross-validation. The Gradient Boosting model outperformed the others, achieving a rsme of 89.7.