Placeholder — fill this in.

Once the archive was consolidated, searching it was still painful. This is about using vision models to auto-tag photos by people, places, and events — so you can actually find the picture you’re thinking of without scrolling through ten thousand thumbnails.