• 

  Home Page

• 

  Upload Page

• 

  Summary of the assignment

• 

  Interactive assessments

• 

  Flashcards

Inspiration?

The idea for AdaptED AI came from seeing how many students struggle to keep up in traditional classroom settings. Not everyone learns at the same pace, and access to personalized tutoring is often expensive or unavailable. Instead of just using AI to give us the answer while learning nothing, I wanted to use AI for an incredible learning experience as it is supposed to be used. We wanted to build something that could bridge this gap—something that could adapt to each learner’s needs and empower them to take charge of their education.

What does it do?

AdaptED AI is a personalized digital tutor that offers real-time, adaptive learning support. It helps students understand difficult concepts through simplified explanations, interactive practice, and feedback tailored to their learning style. Whether it's solving a tough math problem or reviewing a science topic, AdaptED AI acts like a 24/7 study partner.

How did we build it?

Frontend: A responsive web interface built with React.js for an intuitive and engaging user experience. AI Engine: OpenAI’s language models (fine-tuned for educational content) and GEMINI provide explanations, practice questions, and feedback.

Challenges we ran into:

We faced several challenges while building AdaptED AI. One major issue was handling API calls from the backend, especially ensuring they were fast and reliable during high demand. Integrating AI in a way that delivers a tailored learning experience and feedback at each step was another complex hurdle. It required fine-tuning prompts, managing user context, and building logic to make the AI feel like a responsive, intelligent tutor—not just a chatbot.

Accomplishments that we're proud of:

Created a functioning MVP that can tutor users across multiple subjects. Successfully personalized responses based on user input and learning history.

What we learned?

AI can be a powerful tool in education, but thoughtful design is needed to be helpful. Personalization is key. One-size-fits-all solutions don’t work in learning. User feedback is invaluable. Our best improvements came from listening to actual students. We’re especially proud of how we crafted and refined the prompts to personalize the learning experience. Through multiple iterations, we were able to build a system that not only adapts to the user’s style but also identifies and reinforces their weak points. By learning from earlier failures and continuously improving the flow, we created a more effective and intuitive digital tutor experience.

What's next for AdaptED AI?

We intend to make AdaptED AI as diversely capable as possible so it can support learners across a wide range of subjects and fields—from STEM to humanities and beyond. One major upcoming feature is video content integration. We plan to include publicly available educational videos and those provided by colleges, allowing the app to reference specific video segments during tutoring sessions. This will help students better understand concepts by connecting explanations directly to visual, real-world examples.

Complete Presentation: link

Built With

• gemini
• loveable
• react
• shadn
• tailwind
• typescript
• vite