Your Downloads folder is a mess. Your AI shouldn't be.

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

What I Built

Meet LazyDesk AI, a local-first AI assistant designed to handle one of the most boring digital chores: organizing files.

We all have that one folder filled with random screenshots, PDFs, college assignments, receipts, notes, and files named things like final_final_v3_REAL.pdf.

Instead of spending time sorting everything manually, LazyDesk AI lets you select a folder and leave the tedious work to an AI agent.

Here's what it plans to do:

• Smart file sorting: Automatically categorize files into folders such as College, Work, Finance, Images, and Documents.
• Intelligent renaming: Turn confusing filenames into meaningful names based on file contents.
• Quick document summaries: Generate short summaries of PDFs and text documents so you don't have to open every file.
• Duplicate detection: Identify potential duplicates and suggest which copies can be removed.
• Human-approved cleanup: Preview every proposed change before applying it, so the AI doesn't accidentally destroy your carefully accumulated digital chaos.

The goal is simple: spend less time managing files and more time doing literally anything else.

Demo

A demo will be added after the initial prototype is built.

The planned demonstration will start with a deliberately messy Downloads folder and show LazyDesk AI analyzing the files, proposing an organized structure, and sorting them after approval.

Code

GitHub repository: [LazyDesk AI — Repository link to be added]

The project will be open source, with the goal of making local AI-powered file organization accessible to anyone.

How I Built It

The planned architecture combines an open-weight language model with a lightweight desktop interface and a controlled file-management system.

• AI model: Gemma, running locally through Ollama.
• Model inference: Local inference for file classification, meaningful filenames, and document summaries.
• File processing: Python utilities for reading supported file metadata, extracting text, organizing files, and identifying duplicate candidates.
• User interface: A simple desktop interface for selecting folders, previewing proposed changes, and approving operations.
• Safety layer: The AI proposes actions, while deterministic application code validates and executes them. Destructive operations require explicit approval.

The key design decision is to separate intelligence from execution. The model decides what a file appears to be; the application controls what actually happens to it.

Why Does Open Innovation Matter?

A file organizer handles some of our most personal digital information: financial documents, college assignments, private notes, and work files.

Sending all of that information to a remote AI service just to organize a folder feels unnecessary.

Using an open-weight model locally offers three important advantages.

1. Your files stay yours.

File contents can be processed on your own machine instead of being sent to a third-party AI API. Local processing still requires care with logs, temporary files, and other application components, but it gives users much more control over their data.

2. No mandatory per-request AI bill.

Once the model is downloaded, inference can run locally without paying a hosted-model API fee for each request. The trade-offs are hardware requirements, storage, electricity, and processing time.

3. Freedom to experiment.

An open model makes it easier to compare different models, customize prompts, and adapt the classification system to different workflows without tying the entire product to one proprietary API.

For this project, open innovation isn't just a cheaper alternative. It makes a privacy-conscious, potentially offline file organizer practical.

The ideal experience is to set it up once, point it at a messy folder, approve the cleanup, and get back to life.

My Agent Session

A DevRelay agent session will be added if available during development.

The Lazy Payoff

Imagine returning to your Downloads folder after a busy week and finding that every document has a sensible name, related files are grouped together, and important PDFs have summaries ready to read.

Instead of spending 15–20 minutes organizing files manually, the user only needs to review the proposed changes and approve them.

The real measure of success will be how accurately LazyDesk organizes a test folder, how much time it saves, and how reliably it avoids unwanted changes.

Less digital housekeeping. More doing absolutely nothing. 🦥

Prize Categories

• Best Use of Gemma

devchallenge #hf26challenge #hacktoberfest #opensource #ai #productivity