Launch HN: Mosaic (YC W25) – Agentic Video Editing

Mosaic is an agentic video editing platform that uses multimodal AI and a node-based canvas interface to automate video editing workflows. Built by former Tesla engineers, it addresses frustrations with traditional editors by enabling users to create reusable editing agents that can analyze video content and apply intelligent edits through natural language prompts. The platform combines visual intelligence (saliency analysis, object detection, emotion recognition) with a timeline editor and supports export to DaVinci Resolve, Premiere Pro, and Final Cut Pro.

Key Points

• •Node-based canvas UI enables creation of reusable multimodal video editing agents—moving beyond chat-based interfaces that bottleneck workflows for high-volume content creators
• •Multimodal AI analyzes video content including visual concepts, emotions, actions, spoken word, lighting, and shot types to guide intelligent editing decisions
• •Supports parallel branching to A/B test different prompts, models, and workflows, generating multiple variants from the same raw footage
• •Advanced video analysis pipeline includes saliency detection, audio analysis, object significance determination, and optimal safe rectangle calculations for text placement
• •Automates common editing tasks: removing bad takes, repurposing long-form content into clips/shorts, creating montages, optimizing for social platforms, and dubbing with voice cloning
• •Hybrid workflow: agents get content 80-90% complete, then users refine in inline timeline editor and export to professional tools (DaVinci, Premiere, Final Cut)
• •Roadmap includes fully autonomous agents capable of style transfer, self-defined workflows, web asset research, and VFX/lighting modifications without canvas configuration
• •Free tier available with paywall on node runs to cover multimodal AI model costs; API access provided for programmatic integration

Workflow Diagram

Concepts