Project Info

Winner

Crossmint: Best Identity Credential Application Hack (2x rabbit r1 + $1k credit [1st] & 4x Ledger Nano S Plus + $400 credit [2nd])

MindMint: Empowering Education with AI & NFTs

Inspiration

Inspired by the learning incentives offered by Duolingo, and an idea from a real customer (Shray's 9 year old cousin), we wanted to elevate the learning experience by integrating modern technologies, incentivizing students to learn better and teaching them about different school subjects, AI, and NFTs simultaneously.

What it does

It is an educational app, offering two views, Student and Teacher. On Student view, compete with others in your class through a leaderboard by solving questions correctly and earning points. If you get questions wrong, you have the chance to get feedback from Together.ai's Mistral model. Use your points to redeem cool NFT characters and show them off to your peers/classmates in your profile collection! For Teachers, manage students and classes and see how each student is doing.

How we built it

Built using TypeScript, React Native and Expo, it is a quickly deployable mobile app. We also used Together.ai for our AI generated hints and feedback, and CrossMint for verifiable credentials and managing transactions with Stable Diffusion generated NFTs

Challenges we ran into

We had some trouble deciding which AI models to use, but settled on Together.ai's API calls for its ease of use and flexibility. Initially, we wanted to do AI generated questions but understandably, these had some errors so we decided to use AI to provide hints and feedback when a student gets a question wrong. Using CrossMint and creating our stable diffusion NFT marketplace was also challenging, but we are proud of how we successfully incorporated it and allowed each student to manage their wallets and collections in a fun and engaging way.

Accomplishments we're proud of

Using Together.ai and CrossMint for the first time, and implementing numerous features, such as a robust AI helper to help with any missed questions, and allowing users to buy and collect NFTs directly on the app.

What we learned

Learned a lot about NFTs, stable diffusion, how to efficiently prompt AIs, and how to incorporate all of this into an Expo React Native app. Also met a lot of cool people and sponsors at this event and loved our time at TreeHacks!

What's next

Our priority is to incorporate a spaced repetition-styled learning algorithm, similar to what Anki does, to tailor the learning curves of various students and help them understand difficult and challenging concepts efficiently. In the future, we would want to have more subjects and grade levels, and allow the teachers to input questions for the student to solve. Another interesting idea we had was to create a mini real-time interactive game for students to play among themselves, so they can encourage each other to play between themselves.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

FirebaseIn code
JavaScriptIn code
ReactIn code
TypeScriptIn code

4 of 4 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

124 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.