Case Study

From Doing Everything Alone to a YouTube Channel Run from One AI Control Room

Four AI agents that handle the heavy lifting of every video, from the first idea to the finished edit, all run from one dashboard.

← All case studies

The Situation

A YouTuber came to me doing every job on their channel themselves:

  • Guessing at video topics instead of knowing what their audience wanted
  • Planning, scripting, thumbnails and editing all done by hand, one video at a time
  • Editing raw footage taking up the biggest share of their week
  • Ideas, drafts, upload dates and analytics spread across different tools
  • No clear picture of what each video actually cost to make

They wanted to spend their time on camera and on ideas, not on the production line.

What I Built

A control room for the whole channel, with four AI agents doing the heavy lifting. Each agent has one job, and the control room sends them work and keeps track of everything.

The Topic Researcher

Studies the creators in their space, finds the videos that are performing far above normal, and reads what viewers are saying in the comments. It turns that into a list of video ideas the audience is already asking for, instead of guesses.

The Outline Writer

Turns a short brief into a full video plan: a hook for the first few seconds, a reason to keep watching, clear sections, and a strong ending with a next step for the viewer. It follows the rules of what keeps viewers watching, built into every outline.

The Thumbnail Creator

Designs thumbnails from a description and example images, and lets the creator keep refining a design or start fresh until it is right.

The Video Editor

Takes raw footage and turns it into a finished video: it transcribes it, cuts out the mistakes and dead air, cleans up the audio, adds animations and exports the final file. It works for both long-form videos and Shorts, and it remembers the creator's style and past mistakes, so every edit gets closer to how they want it.

The Control Room

One dashboard for the whole channel: an inbox of ideas, a board showing every video's stage, a content calendar, scheduled uploads straight to YouTube, and analytics on watch time, audience retention and subscribers gained.

The Infrastructure

Every agent works on its own and reports back to the control room, so one can be improved or replaced without touching the rest, and new agents can be added as the channel grows. The cost of every AI task is recorded against the video it was for, so the creator knows exactly what each video cost to make. And every agent's access keys are kept separate and private.

The Results

4
AI agents covering ideas, outlines, thumbnails and editing.
1
Control room for the whole channel, from idea to upload to analytics.
2 formats
Long-form videos and Shorts, edited from raw footage.
Every video
Costed, so the creator knows exactly what it took to make.

What Comes Next

The control room is built to grow. New agents, such as one for scripts, descriptions or repurposing videos into social posts, plug into the same dashboard and cost tracking. And every video the editor works on sharpens its sense of the creator's style.