Rafeeq

Inspiration

Rafeeq was inspired by a simple but critical problem: students have access to information, but not to guidance. Many undergraduate students, especially in Algeria, struggle with self-organization, lack of mentorship, scattered resources, and non-personalized learning experiences. Traditional educational platforms focus on content delivery, but ignore learning behavior, collaboration, and real student needs. Rafeeq was born to act as a learning companion, not just a platform — an AI-powered guide that supports students academically, professionally, and collaboratively.

What it does

Rafeeq is an AI-powered educational platform that provides students with personalized learning assistance, smart content generation, collaborative study environments, and intelligent organization tools. It helps students generate quizzes, exercises, schedules, and study materials adapted to their level. Rafeeq also enables virtual classrooms, study groups with video collaboration, shared whiteboards, and centralized resource management. Through its AI hub, Rafeeq acts as a 24/7 tutor, planner, and academic assistant, helping students learn more effectively and consistently.

How we built it

We built Rafeeq using a modern full-stack architecture. The frontend is developed with React, TypeScript, and Vite to ensure performance and scalability. The backend is powered by Node.js and Express, with MongoDB handling data storage. Authentication is secured using JWT and Google OAuth. The AI layer integrates generative AI models and custom agents for tutoring, quiz generation, scheduling, and semantic search using vector embeddings. Real-time features such as study groups, video calls, and collaborative whiteboards are enabled using WebRTC, Socket.io, and PeerJS. The system is designed as modular and scalable to support future expansion.

Challenges we ran into

One of the biggest challenges was designing AI behavior that feels helpful rather than overwhelming. Balancing automation with user control required multiple iterations. Implementing real-time collaboration features such as video calls and shared whiteboards also introduced technical complexity, especially in ensuring stability and synchronization. Another challenge was building an architecture flexible enough to support multiple AI agents while keeping performance and cost under control. Finally, adapting the platform to real student workflows — not idealized ones — required constant validation and rethinking assumptions.

Accomplishments that we're proud of

We successfully built a unified platform that combines AI tutoring, content generation, collaboration, and learning management in one ecosystem. We are particularly proud of the AI Hub, which coordinates multiple specialized agents instead of relying on a single generic assistant. The study group system with real-time video and collaborative whiteboards is another major achievement. Most importantly, we created a solution that directly targets real student pain points, especially for undergraduate students who lack structured academic support.

What we learned

Throughout the project, we learned that education technology is not just about features, but about behavior. Students need clarity, simplicity, and trust in AI systems. We also learned how critical system architecture is when working with AI agents, real-time communication, and scalability. On a team level, we improved our skills in full-stack development, AI integration, system design, and product thinking. We also learned the importance of building for real users rather than theoretical use cases.

What's next for Rafeeq

Next, we plan to enhance Rafeeq’s AI personalization by tracking long-term learning behavior and adapting learning paths more precisely. We aim to expand professional-oriented features such as skill tracking, career guidance, and internship recommendations. Mobile support is a priority to increase accessibility. We also plan to introduce institution-level dashboards for universities and educators, along with multilingual support beyond French. Ultimately, our goal is to position Rafeeq as a complete AI learning companion for students from university to professional life.