AI Video Editing Pipeline

End-to-end agentic pipeline for producing edited, motion-graphic-enhanced video from a raw recording using Claude Code as orchestrator.

Standard stages

1. Raw recording (webcam, screen capture) or HeyGen avatar video input

2. video-use — transcript generation (Whisper or ElevenLabs API) + filler word / silence / retake removal → edited MP4 + word-level timestamp JSON

3. HyperFrames or Remotion — HTML/CSS motion graphics animated to word-level timestamps (animated subtitles, info cards, lower thirds, intro sequences)

4. FFmpeg — final render to MP4

Claude Code orchestrates all steps — writing the HyperFrames HTML compositions, calling video-use, invoking FFmpeg, and iterating on design feedback. ^[src-012]

Key points

• Each iteration improves Claude Code's skills and design philosophy docs for that video type, converging toward drop-in-and-render automation over time ^[src-012]
• Token consumption: ~125K–260K tokens per video editing session ^[src-012]
• Claude Design (web app) can also produce animated video by exporting a site as standalone HTML, then instructing Claude Code to render to MP4 — but cannot natively transcribe audio ^[src-012]
• Word-level timestamps are the critical data dependency: without them, motion graphics fire at wrong moments and feel mechanical ^[src-012]

Related entities

• Hyperframes — preferred motion graphics engine
• Video Use — editing and transcription library
• Remotion — alternative motion graphics engine
• Whisper — transcription backend
• Claude Code — orchestrator

Related concepts

• AI Avatar Content Pipeline — replaces the recording step with HeyGen avatar generation
• Word Level Timestamp Sync — the technical mechanism enabling natural motion graphics
• Agentic Workflows — video editing as an end-to-end agentic workflow

Source references

• ^[src-012] Nate Herk — Video editing & content creation cluster (2026-04-15 to 2026-04-23)