Agriculture still plays a crucial role in the Indian economy by supporting the livelihood of millions of people and maintaining the national food security balance. However, age-old farming techniques are coping with severe challenges like irregular weather patterns, recurrent pest infestations, resource scarcity, and soil fertility. Furthermore, AI and ML enable enhanced real-time perception, data-driven decision-making, and forecasts in both crop and animal husbandry, revolutionizing the industry. To address AI and ML challenges in forecasting sustainable agricultural yields, disease and pest outbreaks, agricultural irrigation, fertilizer application, and livestock health and welfare, the paper offers a consolidated account of AI and ML technologies in every aspect of agriculture. The paper analyzes a broad scope of Machine Learning algorithms from Decision Trees and Random Forest to CNNs and Gradient Boosting to provide a reasoning of their appropriateness for sustainable agricultural tasks. It covers both global advancements and case studies from India, showcasing the work of government agencies, research centers, and sustainable agri-tech startups. This work identifies critical gaps such as insufficient localized datasets, sparse connectivity in rural areas, and high-cost barriers, alongside compiling the relevant datasets, tools, and platforms for AI-based agricultural applications. It also explores future research avenues such as explainable AI, interfacing with IoT devices, privacy-preserving data sharing through federated learning, and automation via robotics and drones. It is clear that the trajectory of AI/ML adoption toward the sustainable agricultural sector is still in developing stages in India, but the potential harnessed through technology to enhance efficiency, sustainability, and resilience in the agricultural sector—both in crop and livestock production—is immense.