Due to the advent of technologies such as ML (Machine Learning) and IoT (Internet of Things), it is feasible to develop solutions to pressing problems in agriculture innovatively, especially when considered in an Indian context. Even though so much progress has been achieved in smart farming, certain critical gaps in precision agriculture concerning optimizing fertilizer management and improving crop health continue to exist. Challenges: Smallholder farmers’ inconsistent usage of IoT technology is due to high cost, connectivity, and lack of accessible and user-friendly platforms. The current detection systems for pests and infections in crops are also not accurate and adaptable for diversified crop types and environments. In this regard, this chapter presents a fully integrated smart agriculture system that uses IoT sensors along with advanced ML approaches to optimize the use of fertilizers, accurately diagnose plant diseases, and generate actionable insights for farmers. The system uses real-time soil and environmental sensor data to precisely recommend fertilizers for balanced nutrient application, and advanced models of image recognition diagnose plant diseases effectively to take corrective measures on time to reduce losses in crops. This empowers farmers to make better-informed choices over irrigation, pest management, and fertilization, by putting together these capabilities within a friendly interface of the Web and mobiles. This approach avoids labor-intensive processes and reduces the environmental footprint by alleviating the misuse of fertilizers and pesticides. Early results declare the system’s capability to increase crop yield significantly, which enhances the quality of the soil and promotes sustainability in farm activities. The model proposed fills existing gaps in smart agriculture, is cost-effective, scalable, and targeted directly to the Indian farmers, which enhance rural economic stability and boost and promote sustainable agricultural productivity.