Final Year Project
University of Sindh • 2025

AI-Powered
Learning Platform

An intelligent LMS featuring automated code grading, an AI mentor chatbot, gamified learning, and real-time analytics — built for modern CS education.

Demo accounts available — sign in to explore as student or educator

5
AI Modules
6
Core Features
15+
API Endpoints
Platform Features

Everything you need to teach & learn smarter

Six integrated modules working together to create an intelligent educational experience.

AI Autograder

Submit code and get instant multi-criteria scoring — correctness, quality, efficiency, and documentation — with detailed test-case breakdowns.

AI Mentor Chatbot

An intelligent assistant powered by Gemini that answers questions, explains concepts, helps debug code, and offers personalized study tips.

Course Management

Create courses, enroll students, post assignments with due dates and rubrics, manage submissions — all from a unified dashboard.

Gamification

Points, badges, leaderboards, and achievement unlocks that keep students motivated and engaged beyond traditional grading.

Analytics Dashboard

Educators see class-wide performance trends and at-risk students. Students track their own progress and growth over time.

Plagiarism Detection

Automated similarity checking across submissions with detailed reports highlighting matching sections and confidence scores.

Key Feature

Autograding that
explains itself

Students submit code, and the engine analyzes structure, runs test cases, evaluates quality, then returns a detailed score breakdown with actionable feedback — in under 3 seconds.

1
Choose a problem and paste your solution
2
Engine runs 10 test cases and analyzes code patterns
3
Get scores across 4 criteria with improvement tips
Try the Autograder
autograder-results.json
{
  "totalScore": 85,
  "breakdown": {
    "correctness":   { score: 45, max: 50 },
    "codeQuality":   { score: 16, max: 25 },
    "efficiency":    { score: 15, max: 15 },
    "documentation": { score:  9, max: 10 }
  },
  "testsPassed": 9 / 10,
  "feedback": "Excellent work! Well-structured and efficient. Add edge-case handling for a perfect score."
}
Technology

Built with a modern stack

Production-grade tools chosen for performance, type safety, and developer experience.

Next.js 15
App Router + Turbopack
TypeScript
End-to-end type safety
Prisma
ORM + SQLite
Gemini API
Google AI models
Tailwind CSS
Utility-first styling
NextAuth.js
Session-based auth
Research Project

University of Sindh
Capstone 2025

This platform was designed and built as a Final Year Capstone Project at the University of Sindh. The research explores how AI-assisted assessment and gamification can improve student engagement in computer science education.

It integrates a multi-provider AI wrapper supporting Google Gemini with a local fallback engine, delivering intelligent autograding and mentoring without requiring expensive infrastructure.

Intelligent Assessment

Multi-criteria code grading scoring correctness, quality, efficiency, and documentation with AI-generated feedback.

Adaptive Mentoring

Context-aware AI chatbot that adapts to student level, provides hints, and explains complex concepts step-by-step.

Engagement Mechanics

Points, XP, badges, and leaderboards that gamify the learning experience and increase retention.

Learning Analytics

Real-time dashboards for tracking student progress, identifying at-risk learners, and measuring outcomes.

Live Demo Available

See it in action

Sign in with demo credentials to explore courses, assignments, autograding, AI chat, badges, and more.