Students often struggle to manage learning materials scattered across PDFs, lecture videos, and notes. We wanted to build a smart study companion that makes learning resources easier to organize, understand, and revise. Vidyora is an AI-powered study companion that processes educational content from multiple sources and converts it into structured, easy-to-use study material. It aims to help students save time, understand concepts better, and prepare more effectively for examinations. We developed Vidyora using a modular approach, with different team members working on separate components. Our work includes PDF text and table extraction, video processing, and converting extracted content into Markdown chunks. These ingestion utilities help prepare educational materials for further AI-powered learning features. We faced challenges with extracting structured information from PDFs, handling different content formats, and processing educational videos reliably. We also had to manage dependencies, test our scripts, and coordinate our work through Git branches and pull requests. We successfully implemented and tested our core ingestion utilities on sample educational materials. Our PDF pipeline processed a 31-page document, our Markdown chunking utility generated structured content, and our video pipeline extracted audio and video frames. We also established a collaborative Git workflow for team development.

We learned how to build modular Python utilities, process different types of educational content, manage project dependencies, test individual components, and collaborate using Git and GitHub pull requests.

We aim to integrate the ingestion pipeline with AI-powered summarization, question generation, personalized revision support, and interactive doubt-solving features. Our long-term goal is to make Vidyora a reliable, personalized learning companion that helps students learn smarter and more efficiently.