A local-first AI curator for a photographer's lifetime archive. It reads your work the way an editor reads a contact sheet, and helps you decide what's worth printing, showing, and keeping.
Every working photographer reaches the same wall. Tens of thousands of frames, years of shooting, and only a handful you ever return to. The rest aren't weak. There has just never been time to find the strong ones.
Culling tools sort by date and rating. Cloud services back things up. Neither one looks at your photographs. Neither tells you which frame is the keeper, which images form a sequence, or which photograph has been sitting in a folder for three years, waiting to be printed.
The work is already there. The seeing is the bottleneck.
Unseen Vision is a local-first curation system for serious photographers. It ingests your archive, evaluates every frame across several dimensions of craft, classifies what it finds, and surfaces what deserves attention: keepers, showcase frames, print candidates, and the sequences hiding in your own archive.
It runs on your machine. Your photographs never leave it, and nothing trains on your work.
It answers the one question the archive keeps asking: what should I do with my photos?
Point it at your archive. It reads every file, pulls metadata and location, and builds a catalog. Your originals are never touched.
Every frame gets a considered read across several dimensions of craft. Not a star rating. The kind of look a good editor gives a sheet of work.
It marks your keepers and showcase frames, flags the edit opportunities, and groups related images into the sequences already in your archive.
Ask in plain language for "my strongest street frames from Tokyo," and it builds the set, lays out the sheet, and suggests where the work could go: a print, a zine, a series, even a show.
You stay the editor. It does the seeing at scale. The judgment stays yours.
Your archive stays on your computer. Nothing is uploaded, nothing trains on your work, and you never need an account to keep your own photographs. The intelligence runs where the photos already live.
Every evaluation is kept and versioned, never overwritten. As the system sharpens, your earlier reads stay intact: a record of how you once saw the work, there to trace. Nothing about your archive is thrown away.
It won't beautify your photos or auto-edit them. It reads the work and helps you decide. The labor of looking at everything goes away. The authorship stays entirely yours.
A considered read of every photograph across several dimensions of craft.
Surfaces your strongest frames automatically, so the best work stops hiding.
Clusters related images into the bodies of work already in your archive.
Ask for what you want in plain language; get a built set back.
Print-ready outputs for zines, prints, posts, and exhibitions.
Flags frames whose potential is one edit away.
Reads location from your files, so you can curate by where you shot.
Collapses the twelve-frame burst down to the one that worked.
A look at Unseen Vision working on a real archive: evaluation, clustering, and curation, start to finish.
Unseen Vision is built for people with real archives: street and documentary photographers, fine-art shooters, anyone with years of frames and a serious relationship to their own work. If you've ever opened a drive of ten thousand photos and closed it again because you didn't know where to start, this was built for you.
It isn't a phone-photo organizer, and it won't prettify your pictures with one click. It assumes you know the difference between a snapshot and a photograph.
Unseen Vision didn't start as a product. It started as the tool its maker needed: a way to read a growing street-photography archive without losing the best work to the sheer size of it. It was built by a working street photographer, for the kind of archive that takes a lifetime to make.
That's the difference. This wasn't designed for a demo. It was built for an archive that mattered, by someone who had to live with the results.
By Xavier
Unseen Vision is in active development, opening to a first group of photographers soon. Leave your email to follow the work, and tell us if you'd like to be considered for beta access.