Quillium

Making Learning More Accessible with AI

Comment

Inspiration

Studying today often means staring at long PDFs, dense notes, and unfamiliar terminology. This can be especially overwhelming for students from underrepresented backgrounds, multilingual learners, and beginners who may not have access to personalized learning tools.

We wanted to build something that makes learning feel less intimidating and more supportive — a tool that adapts to how students actually learn today: interactive, visual, short-form, and accessible.

Quillium was inspired by the idea that education should meet learners where they are, not force everyone into the same rigid study methods.

What it does

Quillium is an AI-powered inclusive learning assistant that transforms static PDFs into interactive learning experiences.

With Quillium, students can:

Upload a PDF (notes, textbooks, slides)

Automatically generate quizzes and flashcards

Translate content into multiple languages

Track learning progress

Create short, AI-generated study videos (“Study Shorts”) for quick revision

Instead of passively reading, learners actively engage with the material in formats that suit their learning style.

How we built it

Quillium is built with a modern full-stack architecture:

Backend: FastAPI (Python)

Extracts text from PDFs using PyMuPDF

Uses AI to generate meaningful multiple-choice questions

Converts questions into flashcards

Supports multilingual translation

Generates short-form explanation scripts and AI voiceovers

Frontend: Next.js + React + Tailwind CSS

Clean, intuitive UI with sections for quizzes, flashcards, progress, and study shorts

Smooth transitions and interactive components

Local state persistence so users don’t lose progress

AI & Media

AI-generated quizzes, summaries, and short-form scripts

Text-to-speech for audio-based learning

Short video previews for quick revision

The entire flow is designed to be simple: upload → learn → revise → track progress.

Challenges we ran into

Extracting clean, structured text from different types of PDFs

Ensuring AI-generated questions were meaningful and not repetitive

Handling multilingual translation reliably

Balancing advanced features with a beginner-friendly UI

Explaining complex content in short, engaging formats for Study Shorts

We addressed these by refining prompts, adding validation layers, and testing with different document types and languages.

Accomplishments that we're proud of

Built an end-to-end learning platform within hackathon constraints

Successfully converted PDFs into quizzes, flashcards, and short videos

Implemented multilingual learning support

Designed an intuitive, visually engaging UI

Added a unique Study Shorts feature that differentiates Quillium from traditional study tools

Most importantly, we created something that feels useful, not just impressive.

What we learned

Simplicity matters more than feature count

AI is most powerful when it supports users, not overwhelms them

Designing for inclusivity improves usability for everyone

Clear UX and storytelling are just as important as technical depth

Short-form learning can significantly improve engagement and retention

What's next for Quillium

Adaptive quizzes based on user performance

Captions and subtitles for Study Shorts

Audio-only learning mode for accessibility

Collaborative study rooms

Deployment at scale for students and educators

Our goal is to continue evolving Quillium into a supportive, inclusive learning companion for students everywhere.

Built With

Share this project:

Updates