Case study · Automation

MUSE: a content department that runs on a schedule

One dashboard that scripts, generates and queues video for six channels across four products. The same machine EXTENSA runs on itself.

The brief

Four products, six channels, one small team. Producing consistent video for all of them by hand is not a discipline problem, it is an arithmetic one. MUSE exists because the maths does not work without automation.

What we built

A pipeline that goes from a product brief to a finished, captioned video: script generation, voiceover, asset assembly, render, then a queue per channel with a dashboard to approve or reject before anything is published.

The approval step is deliberate. Fully autonomous posting is how brands end up with content nobody would have signed off; a human gate costs seconds and prevents that.

  • Brief-to-script generation per product and channel
  • AI voiceover and automated asset assembly
  • Automated render pipeline with captions
  • Per-channel scheduling queue with human approval
  • One dashboard across four products and six channels

Stack

  • Node.js
  • FFmpeg
  • AI APIs
  • ElevenLabs
  • Scheduled workers
Proof of work

Shipped, not mocked up

All selected work →

Want this level of build for your product?

One message with what you are making. You get a fixed scope and a fixed price.

Free 20-minute call · fixed written quote · no discovery fee
Keep reading