Inspiration

Our inspiration comes from the challenges students face worldwide when trying to learn complex STEM subjects. Traditional learning materials are often static, dry, and lack personalization. We wanted to build a smart, conversational learning partner that adapts to every student's native language, lets them learn through natural conversation, and helps them break down tough textbooks instantly.

What it does

OmniLearn AI is an intelligent educational companion that revolutionizes learning. Students can upload their heavy textbooks or notes as PDFs, ask complex educational questions via voice, and get clear, structured answers. It supports multi-subject tutoring (Physics, Coding, Math) in over 100 languages, making quality education universally accessible.

How we built it

We built this entire application utilizing the power of the Baidu MeDo platform. By leveraging MeDo’s robust natural language interface, multi-turn chat capabilities, and advanced full-stack generation, we designed an application that seamlessly orchestrates backend plugins without traditional coding. MeDo allowed us to rapidly integrate file upload capabilities for PDFs and process streaming voice queries to create an interactive user experience.

Challenges we ran into

One of our main challenges was structuring the conversational flow inside MeDo to properly handle diverse multi-turn educational queries while maintaining absolute academic accuracy. Ensuring that the AI references the uploaded PDF accurately without getting confused by formatting took careful prompting and logical iteration within the MeDo editor.

Accomplishments that we're proud of

We are incredibly proud of building a fully functional, production-ready educational app using MeDo in a fraction of the time it would take with traditional coding. Creating an ecosystem that handles heavy academic content and delivers responses in multiple languages smoothly is our greatest milestone.

What we learned

We learned the true power of no-code AI generation through Baidu MeDo. It taught us how to effectively structure AI-to-AI logic, design clean conversational apps, and use multi-turn feedback loops to iterate on features instantly.

What's next for OmniLearn AI

Next, we plan to connect OmniLearn AI with more real-world APIs to provide live quiz rewards, collaborative study rooms for teammates, and specialized training models for regional school boards.

Built With

Share this project:

Updates

Submission history