The latest articles on DEV Community by PERARASU S (@perarasu_perarasus_97283). https://dev.to/perarasu_perarasus_97283 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4182064%2Fd5c36112-a224-4188-8d13-f4a4490c39e6.png https://dev.to/perarasu_perarasus_97283 en PERARASU S Sat, 24 Oct 2026 07:52:36 +0000 https://dev.to/perarasu_perarasus_97283/lazysort-an-ai-powered-file-organizer-23l9 https://dev.to/perarasu_perarasus_97283/lazysort-an-ai-powered-file-organizer-23l9 <h2> <em>LazySort: An AI-Powered File Organizer 🦥</em>* </h2> <h2> The Problem[] </h2> <p>My computer can quickly become cluttered with files such as documents, PDFs, code files, and notes. Organizing them manually takes time and effort.</p> <p>I wanted to explore how AI could help make file organization easier.</p> <h2> My Solution: LazySort </h2> <p>LazySort is a Python-based file organizer built with Streamlit and local AI using Ollama.</p> <p>The goal is to make file organization easier by using AI to help categorize files, instead of relying only on manual sorting.</p> <h2> Technology Stack </h2> <ul> <li> <strong>Python</strong> — application logic</li> <li> <strong>Streamlit</strong> — web interface</li> <li> <strong>Ollama</strong> — running an AI model locally</li> <li> <strong>Qwen2.5 0.5B</strong> — the model I tested locally</li> </ul> <h2> How It Works </h2> <ol> <li>The user interacts with the LazySort interface.</li> <li>The application connects to a locally running Ollama service.</li> <li>The AI model processes a prompt and returns a response.</li> <li>The project aims to use AI-assisted categorization to make file organization easier.</li> </ol> <p>I successfully launched the Streamlit application and tested the local Ollama connection with Qwen2.5 0.5B.</p> <p><em>Note: Add your actual file-sorting test results here after testing the complete workflow.</em></p> <h2> Why Open-Source AI Matters </h2> <p>Using a locally running model gives me more control over how AI is integrated into my project.</p> <p>It also gives me an opportunity to explore AI without depending entirely on a hosted AI API. When configured for local processing, files and prompts can remain on the local machine, depending on the application's implementation.</p> <p>I can also experiment with different models through Ollama.</p> <h2> What I Learned </h2> <p>Building LazySort helped me explore Python, Streamlit, local AI model setup, and communication between an application and an AI service.</p> <p>I also learned that a working AI connection is only one part of the project; the complete file-organization workflow needs to be tested carefully.</p> <h2> What's Next? </h2> <p>I would like to improve the categorization workflow, test it with different file types, improve error handling, and make the application easier to use.</p> <h2> Project Links </h2> <ul> <li> <strong>GitHub Repository:</strong> [<a href="https://github.com/perarasusperarasu0-dev/hoctoberfest.git" rel="noopener noreferrer">https://github.com/perarasusperarasu0-dev/hoctoberfest.git</a>]</li> </ul> <p>Built for the Hacktoberfest Open-Source AI Challenge — Week 3: <strong>Slow and Steady</strong>.</p> <h1> devchallenge #hf26challenge #python #opensource #ai </h1> devchallenge hf26challenge