LazyDesk AI: Let Local AI Tidy Up Your Digital Mess

I built LazyDesk AI, a local-first file organization assistant designed to help people organize their digital workspace with AI-powered category suggestions.

The goal is simple: spend less time sorting files manually and more time doing useful work.

https://lazydesk-ai-etdd2s.v2.appdeploy.ai/

The prototype demonstrates file organization workflows, category suggestions, and manual review before choosing a category.

GitHub Repository: Add your public GitHub repository URL here after uploading the project source code.

LazyDesk AI is a web-based prototype designed around a privacy-conscious, local-first workflow.

The project explores using Ollama with an open-weight model such as Qwen 2.5 to generate file-category suggestions locally. The Ollama connection still needs to be configured and tested on the target computer.

The interface includes a sample workspace, category filters, and editable category suggestions. The current prototype does not actually move or rename files on the user's computer.

Open innovation makes AI more accessible to developers who want to learn, experiment, and build without relying entirely on closed platforms.

Open-weight models and local inference offer an opportunity to explore AI workflows while keeping sensitive file information on a user's own computer, when properly configured.

For me, this project is a step toward making practical AI tools that are accessible, transparent, and privacy-conscious.

The idea behind LazyDesk AI is that organizing digital files should not feel like another full-time job.

My next goal is to validate the local Ollama workflow and explore safe, user-approved file organization on a real computer.

Built for the Slow and Steady — Open-Source AI Challenge.