Abstract— The coffee farming business is a major economic activity in the world's economy. It is considered a highly susceptible crop for nutrient disorders, which are misdiagnosed and cause significant loss in yield and misutilization of fertilizers. This study aims at developing a highly autonomous and accurate diagnosis system for detecting nine different conditions in the leaves of the coffee crop. These nine different conditions include Healthy, Nitrogen (N), Phosphorus (P), Potassium (K), Magnesium (Mg), Boron (B), Manganese (Mn), Calcium (Ca), and Iron (Fe). This research aims at resolving the environmental noise and changes in the images collected from the fields using the CoLeaf-DB dataset. For improving the efficiency and accuracy of the model, advanced ensemble learning approaches have also been used in this study to utilize the potential of different deep learning models in an accurate manner without facing issues of variance and generalization. The methodology for this study includes developing a comprehensive image processing technique for improving the features of the images. For the actual implementation of the proposed system, a high-performance React-based web application has been developed. In addition, the system will include a highly specific Dynamic Multilingual Localization Engine that will support regional languages like Telugu, Hindi, and Kannada. This will bridge the gap between the digital world and the non-technical farming community. Moreover, the system will include an intelligent conversation AI tool like Coffee Assistant, which will be helpful in providing data-driven decisions in terms of nutrient optimization and restorative crop care. The proposed model is an alternative solution in place of costly soil testing in the lab by integrating ensemble modeling, computer vision, and multilingual localization. The proposed model is highly beneficial in the cause of precision agriculture in the cultivation of coffee. KEYWORDS— Coffee nutrient deficiency, Deep learning ensemble, Multilingual localization.