Sceney / Build
Builder
Point it at a live stream. Type what you want in plain English. Sceney, the vision language model (VLM) from Edge Video AI, watches the stream, and the language model underneath builds the app: a panel beside the picture or an overlay on it. Then switch it on live.
Point it at a stream
Choose a live channel or paste a stream link.
Type what you want
Ask what is on screen, or describe the app you want built.
Sceney watches, the language model builds
The VLM reads the stream. The language model underneath builds the app on top of what it sees.
Save it and switch it on
Your app goes to Vision apps, ready to switch on over the live stream.
Two instructions, one stream
Instructions as shown in Edge's explainer film. The second changed the app on the same stream without new code.
What the Builder can and can't change
With a sentence
New views over what Sceney already reads: panels, overlays and on-screen prompts built from categories, keywords, brands and moments.
With engineering
New kinds of output, new places for answers to go, and recognising exact content on TVs, which uses fingerprints rather than instructions.
The facts about the picture always come from Sceney. The language model builds and explains; it never decides what is on screen. Why.