slothos (week 3 DEv)

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

What I Built

I built SlothOS, a local-first AI-powered file organization assistant designed to reduce digital clutter and save time on repetitive file management.

Instead of manually sorting files, searching through folders, and organizing documents one by one, SlothOS helps scan files, categorize them, identify duplicates, suggest better filenames, and organize files into relevant folders. It also includes an undo feature so users can reverse organization actions.

The goal is simple: let AI handle the boring digital housekeeping while you focus on what actually matters.

Demo

Demo Link: https://slothos-yer9.onrender.com/

Run the project locally to explore the dashboard and AI-powered file organization features.

Code

Source Code: https://github.com/rohit060321/slothos

The repository contains the React frontend, FastAPI backend, and integration with a locally running Ollama model.

How I Built It

SlothOS is built using React, Vite, Python, FastAPI, Ollama, and Gemma 3 1B.

• React + Vite: Provides the interactive dashboard.
• FastAPI: Handles backend APIs and file-management operations.
• Ollama + Gemma 3 1B: Enables local AI inference without depending on a paid, cloud-hosted AI API.
• File scanning and categorization: Helps organize files based on their content and characteristics.
• Duplicate detection: Identifies files with matching content.
• Filename suggestions: Helps make file collections easier to understand.
• Undo functionality: Allows users to reverse supported organization actions.

The project follows a local-first approach, keeping AI inference on the user's machine when using the local Ollama setup.

Why Does Open Innovation Matter?

Open innovation made it possible to build an AI-powered productivity tool without relying entirely on expensive, closed AI APIs.

Using open-weight models through Ollama gives developers more control over inference, experimentation, and local execution. It also makes AI experimentation more accessible to students and independent developers who want to build useful tools without needing a large cloud budget.

SlothOS demonstrates how open-source AI can be applied to an everyday problem: reducing the time and effort spent managing digital files.

My Agent Session

I used AI-assisted development to plan, implement, and debug parts of SlothOS, including the React interface, FastAPI integration, and local model workflow.

GitHub Repository

Add your DevRelay agent session link here if you decide to share one.

The Lazy Payoff

SlothOS is built around a simple idea: your computer should spend less time making you manage your files.

By helping with categorization, duplicate detection, filename suggestions, and folder organization, it aims to reduce repetitive manual work and make cluttered directories easier to manage.

The bigger goal is to make digital organization feel less like a chore and more like something that quietly takes care of itself.

Prize Categories

• Open-Source AI Project
• Best use of open-weight AI / local inference, if eligible under the challenge's official category rules