This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend.
What I Built
I built NotePilot AI, a web application that helps students study more efficiently by turning their notes into useful study materials.
Students often spend a lot of time revising lengthy notes and preparing questions or flashcards manually. NotePilot AI makes this process easier by providing three features:
- AI Summarization: Convert lengthy notes into concise summaries.
- Quiz Generation: Generate quizzes from study material to help students test their understanding.
- Flashcard Generation: Turn notes into flashcards for quick revision and active recall.
I built this project with students in mind, especially those who want to spend less time preparing study material and more time learning.
I wanted to build something practical that a friend could use during everyday college study sessions.
Demo
Try the live application:
The application uses a deployed Spring Boot backend hosted on Render.
Code
GitHub repository: NotePilot-AI-2026
The repository contains the frontend, backend, and deployment-related files.
How I Built It
I built NotePilot AI using a simple frontend-and-backend architecture.
Frontend
- HTML
- CSS
- JavaScript
Backend
- Java
- Spring Boot
- REST APIs
AI integration
- Google Gemini API
- Gemini 3.5 Flash Lite
The frontend sends study material to the Spring Boot backend through REST endpoints. The backend communicates with Gemini to generate summaries, quizzes, and flashcards, then returns the results to the frontend.
I deployed the backend on Render and the frontend on Cloudflare Workers.
One important implementation detail is that the Gemini API key stays on the backend rather than being exposed in the browser.
Why Does Open Innovation Matter?
Open innovation matters because developers can learn from existing tools, open-source projects, documentation, and community contributions instead of building everything from scratch.
For NotePilot AI, using an established backend framework and a hosted AI API helped me focus on building useful study features and connecting them into a working application.
The project currently uses Google's hosted Gemini API rather than an open-weight model. Although this means the AI inference depends on an external service and its usage limits, the application's frontend and backend code can be explored in the public GitHub repository.
In the future, supporting an open-weight model could make it possible to offer more control over deployment, privacy, and inference costs.
Top comments (0)