AI Video Editor (V1)
Backend pipeline that transcribes footage and generates AI rough cuts with human-in-the-loop review and export.
The problem
Editing talking-head and tutorial videos manually is repetitive and slow, especially when removing retakes, filler words, and pauses before creating a first cut.
The result
Built an end-to-end AI-assisted editing pipeline that automates transcript-based rough cuts, supports human-in-the-loop review, and exports directly to production formats (MP4 and DaVinci-compatible timelines).
How it works
AI Video Editor V1 is a backend-first editing platform focused on accelerating first-pass video editing. It processes raw videos through audio extraction, Deepgram transcription, and LLM-driven rough-cut generation, then allows users to refine segment validity before export. The system includes project/video lifecycle management, robust JSON persistence, and multiple export targets including preview MP4, high-quality MP4, and OTIO/XML for NLE workflows like DaVinci Resolve.
Tech stack
- Python
- FastAPI
- Deepgram
- LiteLLM
- Groq
- FFmpeg
- OpenTimelineIO
- Pydantic
- Uvicorn
Tags
- LLM rough-cut generation
- human-in-the-loop editing
- word-level transcription alignment
- video export pipeline
- DaVinci Resolve interoperability
- async processing
- REST API
- CLI workflow