Odysser is an AI-powered video editing agent that transforms raw talking-head footage into polished clips with captions, b-roll, split-screen insertions, and motion graphics. It learns a creator's editing style over time, enabling brand-consistent output that can be refined through chat-based instructions. The platform is designed for creators, editors, and marketing teams seeking to produce professional videos in minutes without manual timeline work.
Funding
Funding not disclosed
Founders
Product
Problem
Producing polished video content typically requires hours of manual timeline work in tools like Premiere or DaVinci Resolve, or paying hundreds of dollars per video for a professional editor. Creators and marketing teams often lack the time or budget to add captions, b-roll, split-screen insertions, and motion graphics to raw talking-head footage, which limits their ability to publish engaging content consistently.
Solution
Odysser provides an AI video editing agent that automates the post-production workflow for talking-head videos. Users upload raw footage, review an automatically generated storyboard, and approve captions, b-roll, split-screen insertions, and motion graphics. The system reads the script and visual context before creating insertions and animations, ensuring edits are contextually relevant. It also learns the user's editing style over time, becoming faster and more accurate as edits are approved, rejected, or adjusted. Refinements can be made through chat, eliminating the need to rebuild edits manually in After Effects. The result is a polished, brand-consistent video produced in minutes rather than hours.
Target Audience
Primary users are content creators, video editors, and marketing teams who need to turn raw talking-head footage into polished, publishable clips with captions, b-roll, and motion graphics without manual timeline work.
Features
- Context-aware AI that reads scripts and visual context to generate relevant b-roll, split-screen insertions, and motion graphics
- Storyboard-based review workflow where users approve captions, b-roll, and animations before final rendering
- Style-learning engine that adapts to user preferences through approval and rejection feedback
- Chat-based editing interface for refining generated insertions without manual timeline work
- Brand consistency tools for defining and maintaining captions, titles, b-roll, and motion graphics styles across videos
- Guided AI editing with user approval for most actions, plus automatic split-screen and basic b-roll insertions