StudyLoop: An AI powered productivity planner built for my friend

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

I built StudyLoop, a locally hosted, open-weight-model-powered productivity and planning web app designed to help my friend stay consistent with his work and goals, while also boosting his motivation with AI-generated reports and charts.

Demo

Watch the demo on YouTube

Code

GitHub Repository

How I Built It

StudyLoop uses a locally hosted open-weight LLM (Gemma 3 4B) for AI-powered planning, report generation, and conversational features. The core logic of the app is separated into two parts: the frontend and backend.

The backend is built with Python and FastAPI and handles AI interactions, task management, persistence, and user authentication. The frontend manages the UI and user interactions.

Both parts can be run using a single shell script, making it easier for users to locally host and run the web app.

Why Does Open Innovation Matter?

Open source has been a big part of my life for a long time. Even before getting into tech, I was using GitHub and open-source projects to tinker with games and experiment with different things.

When I transitioned from a commerce background into tech, open source became even more important to me. A huge part of what I know today came from exploring projects made by people who were willing to share their work with the world. It gave me a way to learn, experiment, and build things that I probably wouldn't have been able to otherwise.

So open source isn't just a technology philosophy to me; it has genuinely influenced my journey into tech and has had a positive impact on my life.

For StudyLoop, open innovation made it possible for me to use an open-weight model like Gemma 3 4B locally and build an AI-powered planner around it without relying on a proprietary AI service. I built it to help my friend stay consistent with his goals, keep track of his work, and hopefully become a better version of himself along the way.

In a small way, I'm using the same open ecosystem that helped me get into tech to build something that can help someone else.

My Agent Session

I wasn't able to generate a shareable DevRelay session link, but the project was built with the help of an backboard CLI coding agent during development.

Prize Categories

• Gemma (Used Gemma 3 4B as a local LLM for my project)