MapItOut

Inspiration

The gap in how students were using AI tools, such as ChatGPT, to quickly complete assignments without truly engaging in the learning process inspired us to create MapItOut. As recent graduates from USC, both Yash Sahitya and Sayf Hussain witnessed firsthand how students around us were relying on AI to get things done faster, often without understanding the material deeply. While speed and efficiency are essential, this approach risks bypassing critical cognitive processes necessary for meaningful learning. We wanted to build a tool that strikes a balance—leveraging AI for efficiency while still promoting proactive engagement and critical thinking.

What it does

MapItOut is a dynamic mind mapping tool that guides users through breaking down complex topics by engaging them in conversation with AI. Through a step-by-step process, users are encouraged to interact with an LLM to create personalized mind maps. This approach promotes active learning using the principle of progressive disclosure—where information is revealed progressively to avoid cognitive overload and enhance understanding. The tool not only helps users create visual representations of their ideas but also engages their critical thinking and creativity.

How we built it

We used specialized heuristics with the Langchain library to design the prompt structures that facilitate interactive conversations. For the backend, we built the app using FastAPI, and the frontend was developed with **Next.js. The app is designed to guide users through the process of breaking down long-form text into structured mind maps while allowing room for flexible conversation with the AI. This design ensures a balance between structured learning and creative freedom.

Challenges we ran into

1. *Long-Term Challenge: One of our main concerns was how to create a tool that remains relevant in the long term. As digital intelligence continues to evolve, we don’t want humans to fall behind. Many top researchers believe that human intelligence might be overshadowed by AI in the future. Building tools that keep cognitive processes sharp and ensure people stay engaged with learning is essential to combating this. *MapItOut is our step toward that goal.
2. Experimentation Phase: During the prototyping phase in Google Colab, we struggled to design prompt structures that would enable smooth, seamless conversations between users and the AI. Crafting a flow that was both intuitive and engaging required multiple iterations.
   1. Frontend, Backend, and Deployment: Integrating the frontend and backend and deploying the app smoothly was another challenge. We encountered technical hurdles but managed to overcome them with the help of tools like ChatGPT.

Accomplishments that we're proud of

We are proud of creating a tool that not only addresses a critical gap in how AI is used in education but also promotes proactive learning. MapItOut serves as a step toward using AI in a way that enhances learning without replacing the cognitive processes vital for deep understanding. Building a functioning tool that fosters critical thinking while leveraging AI technology was a significant achievement for us.

What we learned

Throughout this project, we learned the importance of balancing AI assistance with human cognitive engagement. We also gained valuable insights into designing interactive conversational systems, working with the Langchain library, and handling the technical challenges of integrating frontend and backend systems. Additionally, deploying a user-friendly AI tool has given us practical experience in tackling both experimental and production-level challenges.

What's next for MapItOut

In the future, we plan to explore LLM fine-tuning and meta-learning to make MapItOut even more adaptable and personalized. We envision expanding the tool's capabilities to handle a broader range of content types and further enhance its interactive elements. Additionally, we aim to focus on improving scalability and security to ensure the tool can be used by a wider audience. Our next steps also involve refining the user experience, making it more intuitive and accessible for learners and professionals alike.

Built With

• fastapi
• gcp
• gpt
• langchain
• next.js
• python