Agriculture is a major source of livelihood in Pakistan and other agrarian economies, significantly contributing to national economic growth. However, challenges such as excessive herbicide usage, high labor costs, and inefficient weed management continue to limit productivity. This research presents the development of a real-time weed mapping and precision herbicide spraying AgriBot system based on loT and computer vision technologies. The proposed system utilizes a Raspberry Pi 4 (Model B+) integrated with a camera module and deep learning-based object detection to accurately distinguish between crops and weeds. Upon weed detection, the system activates a targeted spraying mechanism to apply herbicide only where required, thereby minimizing chemical consumption and reducing environmental impact. The AgriBot continuously collects field data through integrated sensors and transmits real-time information to a mobile application via loT connectivity, enabling remote monitoring and control. In addition to precision spraying, the system stores farm data for further analysis, supporting data-driven decision-making and optimized farm management strategies. By combining real-time weed mapping, intelligent automation, and remote accessibility, the proposed system enhances operational efficiency, reduces labor dependency, and improves overall crop yield. The integration of lot and computer vision in agricultural robotics represents a significant step toward sustainable, smart, and high-productivity farming systems.