Inspiration
After countless zoom meets that have dragged on too long or ended up being frankly boring, the lack of engagement has become a very common issue, and it time we change that. That's where Engage AI comes in to play.
What it does
Engage AI is a service that provides live transcription to presenters, and uses ChatGPT's AI to generate a multiple choice quiz, where student users score their highest in the spirit of competition, and are verified with the power of verbwire NFTS.
How we built it
- Machine Learning model: ChatGPT AI
- Frontend: JavaScript, TypeScript with React.js
- Backend: Node.js, Express.js
- APIs: React Speech recognition, Socket.IO, Verbwire
Challenges we ran into
- Integrating the Admin Side(Teacher) with the client side(The users)
- Using socket.io to communicate between the servers and client side interfaces
- Verifying if Users own their respective NFTs
Accomplishments that we're proud of
- building a fully working prototype through automation including all tech stacks
- trying new technologies and experimenting along the way
What we learned
- Learned a lot about NFT minting, wallets, and how easily accesible the Verbwire API is
- Learned how to communicate bidirectional messages with Websocket
What's next for Engage AI
- Use MongoDB to store user data in a database
- Find a reliable way to generate MC questions without reliying on ChatGPT
- Use CI/CD pipelines to automate deployment with scripts
Built With
- chatgpt
- javascript
- puppeteer
- react
- socket.io
- typescript
- verbwire
Log in or sign up for Devpost to join the conversation.