Agriculture, the backbone of human sustenance, faces mounting challenges with the global population projected to exceed 9.6 billion by 2050. This increase demands sustainable strategies to boost productivity while addressing environmental constraints and climate variability. In response to these challenges, the system architecture features IoT sensors like NPK analyzers, soil moisture detectors, and temperature monitors, which transmit real-time data via ESP32 and Arduino boards to a centralized database. This data is processed to provide actionable insights through machine learning algorithms. A Random Forest classifier, achieving an accuracy of 99.99%, predicts optimal crops by analyzing soil pH, temperature, and nutrient levels. Similarly, a Convolutional Neural Network (CNN) processes crop leaf images to identify diseases, with an accuracy of 92%. These predictions are displayed on a user-friendly web interface, enabling farmers to access tailored recommendations effortlessly. The fertilizer recommendation module identifies soil nutrient deficiencies and suggests organic solutions and efficient fertilizer application strategies. By promoting sustainable practices, it minimizes environmental impact while maximizing productivity. Moreover, the disease detection module empowers farmers to mitigate crop losses by providing precise diagnoses and actionable treatment options. Data visualization is a pivotal component of Agri Vision, translating raw sensor data into intuitive graphs and charts. This aids in quick decision-making and enhances the user experience. The platform's integration of IoT, machine learning, and real-time data analytics ensures a holistic approach to sustainable farming.