FanumTag
I had over 5,000+ memes and media files with random unsearchable names (like numbers and messy screenshot strings) that were impossible to find. I built this entirely local engine to dynamically load AI models and rename all of my files at once by analyzing what they contained—and it was an absolute success. Version 2.0 is coming soon.
About This Project
Sitting on a hoard of 5,000+ memes and files named with completely random digits or default screenshot strings, making it utterly impossible to search for or find any specific file.
Engineered a lightning-fast batch-renamer that operates entirely offline with zero network access. It dynamically loads required models on the fly to process audio, video, documents, and pictures of all formats, figuring out their content and intelligently renaming them.
An absolute success that flawlessly organized a massive, chaotic media library in record time without ever phoning home. A highly anticipated Version 2.0 is currently in the works.
Backend Developer
2024
Public
Personal
Technology Stack
Project Story
Sitting on a hoard of 5,000+ memes and files named with completely random digits or default screenshot strings, making it utterly impossible to search for or find any specific file.
Engineered a lightning-fast batch-renamer that operates entirely offline with zero network access. It dynamically loads required models on the fly to process audio, video, documents, and pictures of all formats, figuring out their content and intelligently renaming them.
An absolute success that flawlessly organized a massive, chaotic media library in record time without ever phoning home. A highly anticipated Version 2.0 is currently in the works.
Insights & Takeaways
Highlights
- Deep cross-format support: Understands and tags audio, video, documents, and pictures entirely via local inference.
- Dynamically loads specific AI models only when required for maximum speed and minimal resource waste.
Challenges
- Processing over 5,000+ mixed-format files consecutively without causing system crashes or memory leaks.
- Running and managing heavy AI (like SMOLVLM2, Qwen2-VL) on a zero-network, local desktop app constraint.
Lessons Learned
- Gained deep expertise in handling multi-modal AI pipelines in cross-platform desktop architecture via Rust and Tauri.
Related Work
Case Study
Auto Shutdown
Rust + React desktop app to automate PC shutdowns by timer, schedule or idle state.
Case Study
Azkari
Lightweight desktop app that displays Azkar reminders based on time of day — built with Tauri, SolidJS, TypeScript and Rust.
Case Study
EBEC
A full-stack, highly creative web platform built for the ENSIA Business & Entrepreneurship Club, complete with an administration and user management system.
