Design And Implementation Of An Ai-Assisted Final Year Project Management System With Rubric-Based Document Evaluation

The management of final-year academic projects involves proposal submission, supervisor coordination, document exchange, progress monitoring and academic evaluation. When these activities are distributed across separate communication channels, students and faculty may face fragmented records, delayed feedback and limited visibility into project progress. This paper presents the design and implementation of an AI-assisted Final Year Project Management System developed using the MongoDB, Express.js, React.js and Node.js (MERN) stack. The application provides role-specific functionality for students, teachers and administrators, including proposal management, document upload and download, feedback, resource sharing and real-time notifications using Socket.IO. The system also integrates the Grok application programming interface (API), provided by xAI, to support rubric-based academic document evaluation. Submitted documents are assessed against ten predefined criteria with assigned weightages that total 100 marks. The AI module generates criterion-wise marks and improvement suggestions, and students are allowed two AI-assisted evaluation attempts before final submission. The final assessment remains subject to teacher review, and teachers can modify AI-generated marks before finalizing the grade. The paper describes the system architecture, implementation workflow and assessment design. Quantitative validation of grading agreement, scoring reliability and system performance is identified as necessary future evaluation; no such metrics are claimed in this manuscript.