Video intelligence,
on demand.
Cut, composite, caption, and synthesize. Powered by an on-device Rust and WebGPU editing engine, augmented with Google Veo and Gemini generative intelligence — 100% private with zero cloud lock-in.
CORE CAPABILITIES & ARCHITECTURE
LOCAL-FIRST RUST & WGPU ENGINE
The entire editing engine runs directly in the client via WebAssembly and WebGPU. Timeline operations, ripple edits, transforms, and color shaders execute at 60fps with zero server lag.
GOOGLE VEO 3.1 & GEMINI SYNTHESIS
Generate complete high-production videos from scratch. Gemini scripts your scenes, hooks, and narration, while Google Veo generates high-definition video visual beds directly on your timeline tracks.
ON-DEVICE SPEECH & 24KHZ TTS
Auto-captions generated locally via Whisper transformers with zero audio sent to the cloud. Layer synchronized TTS voiceovers with word-level alignment across multi-track audio buses.
FROZEN EDL V1 & SOCIAL EXPORT
Never get locked into a proprietary editing format. Export to standardized EDL v1 JSON, render hardware-accelerated 4K MP4s, or publish directly to Instagram Reels with performance telemetry.
END-TO-END SYNTHESIS PIPELINE
TREND & NICHE ANALYSIS
Extracts high-performing narrative formats, hooks, retention pacing, and sound cues for short-form video engagement.
{
"agent": "trend-analyzer",
"niche": "Tech Hardware Breakdown",
"targetDurationSec": 30,
"retentionStrategy": "high_contrast_hook",
"pacingBpm": 128
}Cut your first scene.
Zero install. Zero lag.
Full multi-track timeline, on-device Whisper captions, GPU shader effects, and Google Veo generative video clips.
Media and project files stay on your machine. IndexedDB storage.
Frozen EDL v1 specification guarantees native platform interoperability.
Modular headless core in packages/editor-core.