News
Sceney goes headless: program the vision language model with Claude and Grok
THE HAGUE - Edge Video AI has made Sceney headless. Sceney's tools are now open to AI assistants through MCP (Model Context Protocol), so Claude and Grok can program the model directly, with no dashboard in between.
Sceney is the vision language model (VLM) from Edge Video AI. It watches live video, reads what's on screen and what's said, and turns it into signals ads and shopping can act on in the moment.
Headless means the assistant is the interface. In plain words, a developer or producer can ask Claude or Grok to:
- add a live HLS stream to a workspace
- watch a span of the stream and say what happened, what was said and who is on screen
- find a moment, like “a close-up of the trophy”
- build an overlay, panel or second-screen app from a brief, using what Sceney sees
- put that app on one channel or on every channel
“Salesforce showed that the next user of software is an agent. Video should work the same way. With Sceney headless, anyone working in Claude or Grok can tell the model what to watch for and what to build, in their own words.” — Joe Ward, CEO and founder, Edge Video AI
About Edge Video AI
Edge Video AI makes Sceney, the vision language model for live video. Edge Video AI is based in The Hague and was founded in 2019.