The Hook
“All you gotta do is just install this Zoah CLI package...”
Product Reveal
“...and then that unlocks so much for you.”
Problem Agitation
“I just like that it gives you this way of not replacing the human designer and not taking away from these traditional design processes.”
Feature Teaser
“So we can do pretty much everything in Zoah just very simple instructions to our agents.”
Feature Teaser
“Doing so with the Zoah MCP server means I can just kind of describe what I want in a conversational-like way over here in the context of my agent session with whatever harness I want to use.”
Feature Teaser
“As the AI model gets better, then the platform becomes that much more valuable because what AI can then do on that platform is just better and better and better too.”
Feature Teaser
“And it just feels natural, like I don't feel awkward doing this... I can brainstorm personas... I can spend as much time iterating here just thinking about the wireframe...”
Feature Teaser
“This is the first time walking through an agentic flow where it's like atomic-based design system, you know? It's building components... You started with persona. This is not stuff that's out there right now.”
Call to Action
“(No spoken dialogue — ambient synthesizer chime outro)”
Synthesizable in Remotion · key components of the cli-agent-canvas-launch template.
Simulates monospace CLI command execution with glowing block cursor and instant response logs.
Use slice on current text string driven by frame count, render monospace font, toggle blinking cursor opacity using frame % 30 < 15.
Rotates flat UI designs in 3D space with subtle camera dolly and slight perspective tilt.
Wrap target UI component in a container with perspective: 1200px, animate rotateX, rotateY, and translateZ with spring().
Sequentially spawns interface cards onto canvas when an MCP tool call finishes.
Map items with an index offset delay: interpolate(frame - i * 4, [0, 10], [0, 1], { extrapolateLeft: 'clamp' }) applied to opacity and translateY.Purpose. Establish product category: MCP bridge from AI agents to design canvases.
Execution. Juxtaposes terminal installation command with immediate high-fidelity UI outputs.
Purpose. Demonstrate that human designers remain in control rather than being replaced.
Execution. Interview clips interwoven with infinite canvas navigation.
Purpose. Prove end-to-end capability from personas to atomic components and live layouts.
Execution. Terminal prompts calling Zoah tools followed by real-time canvas population.
Purpose. Drive conversion to waitlist.
Execution. Minimal title card on dark textured field with website URL.
Objective
Launch video for an AI backend schema generator integrating with Figma.How the recipe adapts
Replace mobile food and sports UIs with backend ERD diagrams and database schema nodes that generate automatically via CLI agent calls.Scene durations (6 scenes, 60s)
Key Remotion components
Introducing Griffin, the first AI to pass the video Turing test
The most emotive and controllable TTS model
We just released QVAC TranscribePsy, a speech-to-text model small enough to run inside a phone app. It covers 25 European languages on a laptop and 8 on a phone, and on an iPhone 17 Pro Max's Ne