What inspired you:
I wanted to make high-quality, personalized tutoring accessible to students in under-resourced communities who lack access to private tutors or after-school help.
What you learned:
I learned how to integrate the Google Gemini API for structured JSON outputs, handle asynchronous AI latency in Express, and build a fully open-source, session-based architecture without requiring user accounts.
How you built your project:
I built the backend with Node.js, Express 5, and MongoDB Atlas, and served a custom, premium vanilla HTML/CSS/JS frontend directly from the Express server, utilizing Gemini 3.5 Flash-Lite for all AI generation.
Challenges you faced:
Debugging unhandled Node.js fetch timeouts that caused the AI interface to spin indefinitely, and wrestling with Mongoose 9's updated pre-save hook syntax.
Built With
- css3
- express.js
- git
- google-gemini
- html5
- javascript
- mongodb
- mongoose
- node.js
- render
- rest-api
- vanilla-js