Treffer: A Full-Stack Implementation of an AI-Powered Chatbot Using Spring Boot and Large Language Models

This paper presents the end-to-end design and implementation of a modern, full-stack AI-powered chatbot that combines the security and modularity of Spring Boot with the user-friendly interface of ReactJS, and the intelligence of Large Language Models (LLMs) accessed through NVIDIA APIs. The system integrates essential web components such as JWT-based authentication, role-based access control, OTP verification, and MongoDB Atlas for data storage. It features real-time AI interaction, secure chat history tracking, and a dedicated admin dashboard for system monitoring. The backend leverages Spring AI SDK to communicate with NVIDIA’s LLMs, while the frontend uses Vite and Bootstrap for fast, responsive UI design. This paper outlines the architecture, methodology, implementation, and testing of the chatbot system, along with challenges encountered during development. Results demonstrate high scalability, performance efficiency, and production-readiness. The project showcases a viable model for AI chatbot deployment in domains such as education, healthcare, customer support, and enterprise IT assistance.