The Hook
“(No spoken dialogue — upbeat percussive synth beat)”
Problem Agitation
“(No spoken dialogue — rhythmic keyboard typing sound effects)”
Product Reveal
“(No spoken dialogue — message send and receive pops)”
Feature Teaser
“(No spoken dialogue — ambient digital click pulses)”
Feature Teaser
“(No spoken dialogue — whoosh and confirmation chimes)”
Product Reveal
“(No spoken dialogue — incoming message chimes and typing sounds)”
Feature Teaser
“(No spoken dialogue — percussion accents)”
Call to Action
“(No spoken dialogue — final resolving chord)”
Synthesizable in Remotion · key components of the agent-chat-orchestration template.
Simulates smooth iOS chat progression with automated viewport shifting as new bubbles appear.
Use Remotion interpolate on frame counter with spring physics to translateY a container component as new chat messages are rendered.
Background phone numbers scroll rapidly and blur while brand pills snap into checkmarked slots.
Layer infinite repeating numeric grid moving via translateY while animating opacity and scale of pill components using spring configs.
Words swap in-place with spring bounce and inline graphics.
Animate span elements along X/Y axes with quick exit translations and staggered spring entry for new tokens.
Purpose. Frame high-friction user request instantly.
Execution. Typing message bubble in high-key white space.
Purpose. Demonstrate autonomous reasoning and service discovery.
Execution. Rapid phone book search matrix resolving into recognized service icons.
Purpose. Show full problem resolution without app switching.
Execution. Multi-agent iMessage thread completing lodging, transit, and dinner reservations.
Purpose. Direct users to beta onboarding.
Execution. Crisp centered URL card.
Objective
AI travel concierge booking a weekend getaway via WhatsApp.How the recipe adapts
Replace iMessage skin with WhatsApp UI elements; query flight, hotel, and car rental service agents instead of offsite logistics.Scene durations (7 scenes, 45s)
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