Sort a folder of screenshots and images by what they contain
Run OCR on each image, then have Jev pick a category from the extracted text so a messy image folder gets filed.
Author's figure: Author's figure: roughly 900 images, 40 s (Fayaz Ahmed). Reported by the builder, not measured by us.
The use case
Marketers, support teams and researchers collect folders of screenshots, scanned notes and saved images with useless file names. This pattern runs OCR on each image to get its text, then asks Jev a Choice question over the categories you care about, such as receipt, chat screenshot, chart, slide or product photo. Files move into matching folders or get tagged. Images with little text are better handled by Jev's image preview or a vision model, and someone glances over each folder before deleting or sharing anything.
Example workflow
- Point a script at the image folder
- Extract text with Apple Vision OCR or Tesseract
- Ask Jev a Choice question over your categories
- Move or tag files by category in Google Drive or Finder
- Review each folder and fix mislabels
Tools
- Jev (TypeSafe AI)
- Tesseract OCR
- Google Drive
Risks and guardrails
- Images without text cannot be sorted by OCR alone
- Private screenshots may hold sensitive data
- Moving files automatically can break links
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- 900 images in 40 seconds by Fayaz Ahmed on Made with Jev: madewithjev.com/builds/ocr-image-classifier
- 900 images in 40 seconds: original post or repo by Fayaz Ahmed: x.com/fayazara/status/2100953838891192789
- Google Drive (official page): developers.google.com/workspace/drive (checked 2026-11-19)