🧩 Project Story

About the Project

🌟 Inspiration

This project was inspired by the need for a smarter, more interactive way for students to study from PDFs. I wanted to build an AI-powered study companion that could read coursebooks, generate quizzes, explain answers, and simplify the entire learning experience.

🛠️ How I Built It

• Implemented a RAG pipeline (chunking + embeddings) to enable contextual, cited chat responses.
  
• Designed a split-view PDF reader so users can read while interacting with an AI tutor.
  
• Built a Quiz Engine that generates MCQs, SAQs, and LAQs, evaluates answers, and provides explanations.
  
• Added a progress dashboard that visualizes performance and learning trends.
  
• Created a ChatGPT-style responsive UI for smooth cross-device interaction.
  
• Integrated YouTube video recommendations based on PDF context.

🎯 What I Learned

• How to build end-to-end Retrieval-Augmented Generation (RAG) systems.
  
• Techniques for efficient PDF parsing, chunking, and multi-file handling.
  
• Designing clean, responsive full-stack interfaces.
  
• Balancing performance, accuracy, and scalability in AI-driven features.
  
• Improving prompt engineering and understanding model behavior.

⚠️ Challenges I Faced

• Handling large PDFs while keeping the UI responsive.
  
• Ensuring high-quality, diverse quiz generation.
  
• Maintaining RAG accuracy with PDFs containing complex layouts or formatting.
  
• Integrating multiple advanced components into a cohesive, intuitive UX.

✨ Final Thoughts

This project was a rewarding experience that allowed me to blend full-stack development with applied AI. It strengthened my skills in building intelligent, user-centered applications, and I’m excited to continue improving and expanding it.

Built With

• gemini
• javascript
• llm
• react