I got tired of trying to find one specific photo on my PC and realizing Windows Search was basically useless for image content.
So I started building my own solution.
The idea is simple:
• Runs fully local - an AI model scans your images and generates tags based on what’s actually in the photo (objects, scenes, activities, etc.)
• Search photos using natural language instead of filenames and folders
• everything stays on your device
• Designed to work on older and lower-end hardware, not just machines with powerful GPUs
• Building a Spotlight-style quick search overlay so you can instantly find images without opening the main app
• Planning to write AI-generated tags directly into image metadata, so if you ever stop using the app, the tags stay with your files
It’s still early and rough around the edges, but I wanted to share it before locking in the beta.
A few questions:
• Would you actually use something like this? • What would make it a must-have for you? • What’s the most frustrating thing about managing large photo libraries today? • Any local-AI photo tools you’ve tried that got something wrong?