What I Built

For the Hacktoberfest Weekend Challenge: Build for a Friend, I built StudyNote AI — a small local AI study assistant designed to help a student turn long study notes into useful revision material.

The idea is simple:

Paste study notes → Get a simple summary + important questions + short answers.

The project uses Gemma 3 1B through Ollama, so the AI runs locally on the user's computer instead of relying on a paid cloud AI API.

Who I Built It For

I built this for a friend who spends a lot of time going through college notes before exams and creating revision material manually.

The problem wasn't that they couldn't understand the notes. The problem was the amount of time spent converting those notes into something easier to revise.

I wanted to make that process faster while keeping the project small and practical.

The Problem

When preparing for exams, students often have:

• Long class notes
• Multiple topics to revise
• Limited time before exams
• Difficulty identifying the most important points

Using a general AI website can help, but it can also mean sending personal study material to an external service.

So I wanted to try a different approach.

What if the AI could run locally?

How StudyNote AI Works

The application has a very small architecture:

Student
   ↓
Study Notes
   ↓
Flask Web App
   ↓
Ollama
   ↓
Gemma 3 1B
   ↓
Summary + Questions + Answers
   ↓
Student

The user pastes their notes into the web application.

Flask sends those notes to the locally running Ollama server.

Ollama runs Gemma 3 1B and generates:

1. A simple summary
2. Three important questions
3. Short answers

The generated result is then displayed in the browser.

Demo

The working application looks like this:

The basic workflow is:

1. Paste notes
2. Click Generate Study Material
3. Wait for Gemma to process the notes
4. Review the generated revision material

GitHub

The complete source code is available here:

https://github.com/25A35A0531/StudyNote-AI

How I Built It

I kept the project intentionally small.

Technologies

• Python
• Flask
• HTML/CSS
• Requests
• Ollama
• Gemma 3 1B

The Flask application communicates with Ollama using its local API.

I didn't use a database, authentication system, paid AI API, or complicated frontend framework because the goal was to build something useful within the weekend.

Why Gemma and Ollama?

The most important part of this project is that the AI is not just an optional feature.

Gemma is what actually generates the study material.

I downloaded Gemma 3 1B using Ollama and ran it locally on my laptop.

This made the project possible without purchasing an AI API or creating an API key.

Why Open Innovation Matters

This project made me appreciate one of the biggest advantages of open AI models: access and control.

With a local open model, I can experiment without depending completely on a paid API.

For a small student project, that matters.

The application can run on a personal computer with the model downloaded locally. Study notes don't need to be sent to a separate AI provider just to generate a summary.

It also means I can experiment with different local models in the future instead of rebuilding the entire application around one closed API.

For this project, open AI wasn't just a technology choice.

It shaped the way I designed the application.

What I Learned

This project was small, but I learned a lot from building it.

I learned how to:

• Run an open model locally with Ollama
• Communicate with a local AI model from Python
• Build a simple Flask application
• Design a useful AI workflow around a real user problem
• Connect a local AI model to a web interface
• Package and document a project for others to run

Built for a Friend

The most important part of this challenge for me was not simply building an AI demo.

It was building something for a real person.

After building the first version, I shared it with my friend to see whether it actually helped with studying.

What's Next?

This is currently a small MVP, but there are several things I could add later:

• Multiple-choice quiz generation
• Difficulty selection
• Custom number of questions
• Copy/download generated material
• Quiz mode
• Study history
• Better UI

For this challenge, I decided to keep the first version small and focus on solving one problem well.

Prize Category

I am also submitting StudyNote AI for the:

Best Use of Gemma

because Gemma 3 1B is the core AI model responsible for generating the study material.

Final Thoughts

StudyNote AI started with a simple question:

Can I build something small with open AI that would actually help someone I know?

The answer was yes.

It isn't a huge application, and that was intentional.

It is a small tool that turns study notes into revision material using an open model running locally.

For me, that's what made this challenge interesting: building something that is not just an AI demo, but something connected to a real person's everyday problem.

Thanks for reading! 🚀

Hacktoberfest #AI #OpenSource #Gemma