To address food security and sustainability in the quickly changing world of agriculture, crop yield and fertiliser consumption optimisation are essential. Using machine learning techniques, we created an integrated crop and fertiliser recommendation system aimed at improving agricultural output. The Crop Recommendation System and the Fertiliser Recommendation System are the two primary divisions of the system. The Random Forest approach, which was chosen after a thorough comparison with various machine learning algorithms like Decision Trees, Support Vector Machines (SVM), K-Nearest Neighbours (KNN), and Native Bayes, is utilised by both components. We conducted a thorough evaluation procedure using accuracy, precision, recall, and F1-score as metrics, and found that the Random Forest algorithm consistently produced the best results in terms of resilience and accuracy. In order to suggest the best crops for a particular plot, the Crop Recommendation System takes into account a number of important factors, such as rainfall, temperature, humidity, pH, and soil nutrients (nitrogen, phosphorus, and potassium).