Inspiration
Online examinations have become common, but ensuring fairness, preventing cheating, and reducing teachers' workload remain major challenges. We wanted to build a platform that combines AI-powered proctoring with an intelligent teacher assistant to make online exams secure, efficient, and easy to manage. Our goal was to create a solution that not only detects suspicious activities but also helps educators analyze exam data and generate insights using AI.
What it does
Alice Exam Proctor is an AI-powered online examination and proctoring platform that provides:
🎥 Real-time webcam monitoring during exams. 👁️ AI-based face detection and suspicious activity detection. 📸 Automatic screenshot capture when suspicious behavior is detected. 📊 Teacher dashboard for managing exams and monitoring students. 🤖 Gemini-powered AI chatbot that assists teachers by answering questions, explaining results, generating reports, and providing exam-related insights. 📈 Student performance analysis and result visualization. 🔒 Secure authentication and exam management. How we built it
We built Alice Exam Proctor using a modern full-stack architecture:
Frontend
React.js Tailwind CSS Chart.js
Backend
Django Django REST Framework WebSockets for real-time communication
Database
MongoDB Atlas
AI & Computer Vision
MediaPipe Face Mesh face-api.js Google Gemini API for the AI teacher assistant
Deployment
Vercel (Frontend) Render (Backend) Challenges we ran into Implementing real-time AI monitoring with low latency. Synchronizing webcam events using WebSockets. Handling deployment issues such as CORS and API integration. Optimizing AI detection to reduce false positives. Securing API keys and backend services while integrating Gemini. Ensuring the platform remained responsive during live examinations. Accomplishments that we're proud of Successfully built a complete AI-powered online proctoring platform. Integrated a Gemini-powered chatbot into the Teacher Dashboard. Achieved real-time communication between students and teachers. Developed automated suspicious activity detection. Created an intuitive dashboard with performance analytics. Built a scalable architecture suitable for educational institutions. What we learned
Throughout this project, we gained practical experience with:
AI integration using Google Gemini API. Real-time communication using WebSockets. Computer vision with MediaPipe and face-api.js. Full-stack development using React and Django. Deployment, API security, and cloud hosting. Designing AI-assisted educational applications. What's next for Alice Exam Proctor
Our future roadmap includes:
🎙️ Voice-enabled AI assistant for teachers. 📱 Mobile application for students and educators. 📝 AI-generated question papers and quizzes. 📄 Automatic PDF and Excel report generation. 🌐 Multi-language support. ☁️ Cloud scalability for large-scale examinations. 📊 Advanced analytics with predictive insights. 🤖 Personalized AI feedback for every student.
Log in or sign up for Devpost to join the conversation.