Betel farming plays a vital role in Sri Lanka’s agricultural sector. However, traditional methods and limited access to modern technology hinder productivity and sustainability, especially for new and small-scale farmers. This paper presents BetelCare, a machine learning-powered mobile application de veloped to support betel farmers, particularly newcomers—by offering real-time insights, predictive analytics, and interactive assistance through a chatbot interface. The system incorporates a harvest prediction model, weather-based cultivation recommendations, and disease detection using convolutional neural networks (CNNs), along with personalized treatment suggestions. A key feature of the system is the integrated chatbot, which guides users through prediction workflows—such as demand forecasting or market price estimation—and responds to agricultural queries using a retrieval-augmented generation (RAG) approach. The chatbot enhances accessibility, especially for non-technical users. The system was implemented using a cross-platform mobile application built in Flutter, machine learning models trained in Python via Google Colab, and a Flask backend. Additional features like geolocation-based disease mapping and predictive weather impact models support data-driven farming decisions. Experimental results confirm the system’s effectiveness in ad dressing key challenges faced by betel farmers in Sri Lanka.