AI Slop or Automating the Grind?

Description and use

An automated pipeline for YouTube video creation using Claude, Remotion, ElevenLabs, and Whisper to reduce manual effort and speed up production.

Creating YouTube videos was time‑consuming and costly, with low viewership despite decent quality.

The pipeline reduces manual effort, speeds up production, and allows scaling of YouTube video creation.

1. Concept module: generate and rank script concepts with Claude and Jev.
2. Research module: gather evidence from public domain archives and academic sources, producing a source document.
3. Footage module: identify and download relevant footage, index clips.
4. Script module: write core script and refine with Claude.
5. Editing module: render maps, stitch video parts, add voice and music (ElevenLabs clone).
6. Prep module: create thumbnails, subtitles, translate subtitles into 10 languages using Whisper.
7. Publish to YouTube and monitor metrics.

I threw them at Claude, it digs up similar stuff on its own and fleshes out the concepts.

The footage module ... tens of gigabytes of video get downloaded, cut into clips, and their indexes written to a file for later use.

For the voice, I went with ElevenLabs and cloned my own.