The critical part of agriculture for global food security requires efficient decision-making tools to optimally the crops to grow and the fertilizer to use. We can analyze artificial intelligence techniques such as machine learning models to improve predictive capabilities. Now by using the benefits of the machine learning and deep learning, the system can advise, real-time and personalized, crop selection and fertilizer application recommendations as a function of the environmental & soil conditions. To gather insights, the system utilizes a dataset from the Kaggle which is very comprehensive and includes information about the critical agronomic factors such as levels of nitrogen, potassium and phosphorus, temperature, humidity, rainfall, pH value, and soil type. After that, we trained and evaluated XGBoost, logistic regression, KNN, decision tree, and gradient boosting. XGBoost algorithm was outstanding with an accuracy rate of 98.50%.