What I Built

I built LaterLess — Let Your Local AI Handle the Busywork.

I wanted to build something practical with AI that could help with everyday tasks. Organizing files manually can be annoying, especially when there are lots of files with unclear names.

LaterLess uses AI to suggest suitable categories and cleaner filenames. I can review the suggestions, select the changes I want, and apply them. It also has an undo feature to reverse the last organization operation.

I tested it using sample files and successfully tried the complete process, from scanning files to organizing a selected file and undoing the changes.

My goal is simple: spend less time organizing files and more time doing other things!

Demo

LaterLess currently runs locally on my computer.

I'll share a short video demo showing how it scans files, generates AI suggestions, organizes selected files, and undoes the changes.

Code

GitHub Repository:

25A35A0530 / LaterLess

LaterLess — Let Your Local AI Handle the Busywork

LaterLess is a local AI-powered file organizer that helps reduce the time spent organizing files manually. It uses an open-source AI model through Ollama to suggest meaningful file categories and cleaner filenames, allowing users to review and approve changes before applying them.

Built for the Hacktoberfest Open-Source AI Challenge — Week 3: Slow and Steady.

✨ Features

• AI-powered suggestions: Uses a locally running AI model to suggest file categories and cleaner filenames.
• File scanning: Scans files in the selected sample folder.
• Preview before applying: Lets users review suggestions before making changes.
• Selective organization: Allows users to choose which suggested changes to apply.
• Automatic organization: Moves approved files into category folders and renames them.
• Undo support: Provides an option to reverse the last organization operation.
• Local AI processing: Uses Ollama to run the AI model locally.
• Simple web interface: Provides an…

View on GitHub

How I Built It

I built LaterLess using Python for the backend and HTML, CSS, and JavaScript for the web interface.

For the AI, I used Ollama with the Qwen2.5-Coder 3B model, which runs locally on my computer.

The application scans sample files, asks the AI for category and filename suggestions, and lets me choose which changes to apply. Once approved, the application moves and renames the selected files. The undo feature helps restore the previous state.

Building this project helped me understand how to connect a local AI model to a working application instead of just using AI to generate answers.

Why Does Open Innovation Matter?

As a student, open-source tools give me the opportunity to experiment, learn, and build projects without depending entirely on paid AI services.

Using Ollama and an open-weight model allowed me to explore local AI inference and connect it to my own application. I could test the idea on my computer and learn by making improvements along the way.

I also like that other developers can explore the code, suggest improvements, and build on the idea.