VidCutAI
AI-Assisted Podcast & Multicam Video Editor
Role
Technical Founder & Lead Developer
Status
Core functionality complete, undergoing real-world testing and iteration toward a stable release
Stack
The Challenge
Multicam and podcast editing is slow and repetitive. Syncing media, managing multiple sources, tracking speakers, and assembling a clean timeline takes hours. Many AI tools require full uploads, are expensive, or don’t fit real post-production workflows.
The Solution
I built a desktop-first editor with a robust media pipeline and AI-assisted analysis. FFmpeg handles conversion and normalization, AI services handle transcription and diarization, and the app structures speakers and segments into an edit timeline that can be exported into NLE-friendly formats for finishing.
Key Features
- Hybrid Processing (Local pre/post + API-based AI)
- Automated Speaker Diarization and Transcription
- Multicam Source Management (multiple angles, multiple video sources)
- Timeline Export (Final Cut Pro XML, DaVinci Resolve workflows)
- Speaker-to-Angle Mapping for Structured Edits
Description
A desktop application that streamlines podcast and multicam editing. It combines local pre and post processing with API-based AI workflows to transcribe, diarize, and generate structured edits, exporting timelines for professional NLEs.
