artgraph.co · A system for abstract art discovery
ArtGraph
The first dimensional discovery system for abstract art — the infrastructure layer the art world has been missing.
Not curation. Encounter.
How do you give language to something that inherently has none?
That's the problem at the center of abstract art — and the reason every existing system for finding it fails. Galleries curate for convincing, not for encounter. Search engines categorize by style, as if a painting were a product on a shelf. Platforms optimize for inventory, not for the person standing in front of something that won't let them go.
I bought a painting once, walking through a small street shop in Udaipur. No gallery, no lighting, no one explaining why I should want it. I was simply compelled. That painting is still in the tube I carried it home in — because the infrastructure to place it correctly in my life doesn't exist.
Abstract paintings have a 1-to-1 relationship with the viewer. Not a demographic. Not a taste profile. One painting, one person, one encounter that can't be manufactured.
ArtGraph was built to make that match possible. Nine dimensions. Real sentiment. The rough shape of what you're creating in your space, mostly not yet said in language. And finally — how it makes you feel, mapped against how it has made others feel.
The Taxonomy of Abstraction
A nine-dimension scoring framework — the first formal language for what abstract paintings are doing beneath the surface. Applied to a catalog of 252 works. The infrastructure layer.
The Abstraction Engine
A live conversational AI that translates a person's life, mood, and space into a specific painting. The proof of concept. The Discover Weekly of abstract art — built on ArtGraph the way Spotify built Discover Weekly on the Audio Graph.
The Book
ArtGraph: A Discovery System for Abstract Art
The Taxonomy of Abstraction and the Future of How We Find Art
Publisher conversations underway.
The Vision
ArtGraph as licensable infrastructure for Christie's, Sotheby's, Artsy — the translation layer between a person's life and a painting that those platforms have never had.
A founding engineer of Facebook used the Abstraction Engine — a 12-minute conversation about his life in Venice Beach. He found his painting. He began sharing it with his network. Two independent love signals in 22 minutes.
Why every existing system fails
Spotify built the Audio Graph — a proprietary dimensional map of music. From it came Discover Weekly: the most precise recommendation engine in cultural history. Spotify didn't just organize music. It gave music a language it could be searched by. ArtGraph is built on the same principle.
But there is a difference worth naming. Spotify organized what music already knew about itself — tempo, key, time signature, genre. The Audio Graph made existing vocabulary computable. What the Taxonomy of Abstraction had to do is harder: invent the language from scratch, for a medium that has none.
The dating site parallel — and where it ends
Dating platforms are the closest human model for 1-to-1 dimensional matching where the match is deeply personal and consequential. They solved that problem at scale. But both sides of that match have language. A person can describe themselves. The other entity can do the same.
Abstract painting cannot speak. It has no representational anchor — no face, no place, no story you can name before you've felt it. A portrait tells you who. A landscape tells you where. Abstract painting tells you nothing, on purpose. That is its power. That is also why every existing system for finding it fails.
The conflict of interest
Curators write wall texts. Art advisors make recommendations. Gallerists tell you what a work means and why you should want it. They are translating — or attempting to. But they are translating in the service of a sale. The bias is structural, not personal. A curator champions what they've chosen to show. An art advisor earns a commission on what moves. A gallerist needs the wall to clear.
ArtGraph has no inventory to move. No wall to clear. No commission riding on which painting you choose. The Taxonomy only knows the nine dimensions — and which painting scores closest to what you're actually looking for. Nobody with a stake in the outcome was ever going to build this.
A 252-painting catalog. A conversational AI. Your life, your space, your painting.
Find your painting →The unified theory behind the system. Publisher conversations underway.
Request the proposal →Press, publishers, collectors, collaborators, investors. I'd welcome a conversation.
ritu@rituart.com →Ritu Raj is an emerging contemporary abstract painter based in Phoenix, Arizona, represented by Jarrow & Goodman in Los Angeles. He maintains a catalog of 252 original paintings across eight collections — and built ArtGraph because no infrastructure existed to help people find the right one.
Before returning to painting full-time in 2020, he was a serial entrepreneur who built companies that created entirely new categories.
Painting is a return to heritage. He grew up in New Delhi as the son of K.B. Goel, one of India's preeminent art critics — a man who championed the generation that defined modern Indian art. Souza, Husain, Swaminathan were not names in a textbook. They were the people in the room. ArtGraph carries that inheritance forward: the belief that abstract art has always deserved a language precise enough to find the person it belongs to.