Inspiration Traditional learning tools often feel static and linear—one answer at a time, one way of thinking. We wanted to create something more organic, something that mirrors how real understanding branches out. Inspired by how people brainstorm on whiteboards and how ideas connect in non-linear ways, we built ThinkBoard: a platform that visualizes the learning process as a branching, customizable dialogue with an AI tutor.

What it does ThinkBoard allows users to ask a question and receive dynamic, visual responses in the form of an expanding tree. Each node represents a follow-up explanation, question, or alternative perspective, helping users break down complex ideas. The AI tutor's personality and teaching style can be customized—whether you want concise definitions, Socratic questioning, or even humor.

How we built it Frontend: Built with React and type script to create an interactive and intuitive tree structure that grows as the conversation evolves. Backend: Node.js + Express server handles API requests and manages user sessions. AI Integration: OpenAI's API powers the conversational engine, tailored with custom prompts based on user preferences. Customization: We store user-selected tutor styles and learning modes, which influence the way OpenAI responds. Challenges we ran into Designing a tree structure that remains readable and engaging as the conversation grows. Managing prompt engineering to adapt to different tutor "personalities." Ensuring the experience stays fast and responsive as more branches are added. Finding the balance between too much and too little information in the AI's replies.

Accomplishments that we're proud of Creating a seamless experience that makes learning feel visual, adaptive, and personal. Allowing users to actively explore knowledge paths rather than passively receive information. Making AI feel less robotic and more like a real tutor you can customize to your learning style.

What we learned How to work with OpenAI's API in a dynamic, user-directed context. How important good UI/UX is when trying to display complex, growing information. That AI is powerful, but how you frame the questions (and answers) is what makes or breaks a great learning experience.

Built With

Share this project:

Updates

posted an update

Future of ThinkBoard:

Implementing images and figures can further develop the user’s understanding. Additionally, images could be used as a form of input. Having lines that connect the boards together increases the readability of the workspace as a whole Sharing boards with collaborators not only helps students learn topics quicker but also increases the userbase.

Log in or sign up for Devpost to join the conversation.