Agriculture plays an essential role for securing global food supply, and in so, crop yield and fertilizer use have to be optimized. To answer this growing demand for food, it will be necessary to increase the Crop Yield to its optimal possibility. This academic analysis is fundamentally centered on the application of machine learning (ML) frameworks to derive highly accurate and data-supported recommendations regarding crop selection for cultivation and fertilizer usage. The analytical framework delineates the most appropriate crops alongside the requisite quantities of fertilizers for diverse geographical regions; this assessment incorporates a multitude of variables including concentrations of nitrogen, phosphorus, and potassium, as well as humidity, temperature, precipitation, pH levels, and classifications of soil present within the dataset. The models that were employed in the analytical process encompass: XGBoost, Random Forest, Logistic Regression, and Gradient Boosting, all of which were trained utilizing the specified dataset. Subsequent to the evaluation of each model’s efficacy, it has been determined that the Random Forest algorithm exhibited superior performance relative to all other algorithms, achieving an exceptional accuracy rate of 99.99%. This model was then chosen for real-time predictive analytics to generate crop and fertilizer recommendations. The application architecture delineates two primary user roles: For administrators and general users. User authentication, dataset oversight and data preprocessing tasks are handled by administrators, while general users, the farmers, register through OTP and access their personalized dashboards. With these dashboards, users derive actionable insights on the most suitable crops to grow and requisite levels of fertilizer requirements that are relevant as per the information on returned agricultural schemes offered by the government.