A Secure and Personalized AI Chatbot for School Performance Monitoring using Local LLM and MongoDB

Traditional School Management Systems often provide static information, creating a communication gap for parents seeking real-time updates on their child's academic progress. While commercial AI chatbots offer a solution, their reliance on third-party APIs introduces significant data privacy risks and recurring operational costs, making them unsuitable for many educational institutions. This paper addresses these challenges by proposing a secure, personalized, and self-hosted AI chatbot designed to empower parents with instant access to their child's performance data. The system leverages a locally deployed Large Language Model (LLM), such as Llama 3–8B Instruct, served via Ollama and integrated with a MongoDB database, ensuring that all student data remains within the institution's private infrastructure. A secure authentication mechanism based on a unique roll number and a secret code ensures data isolation and privacy for each parent. By eliminating dependency on external APIs, this framework offers a cost-effective, secure, and scalable solution that enhances parent-school engagement while maintaining complete data sovereignty.