— Loading
Back to all case studies
— AI Content·Distributed (US + EU)·May 10, 2026·11 days
Multi-channel SaaS launch (anonymized)
Daily voice-locked content pack across 5 channels, generated by 1 script and AI-tell-linted before render.
— Manual content QA / day
Before: 3 hrs
After: 0
-100%
— Channels covered by 1 script
Before: 1
After: 5
5x reach
— AI-tell phrases shipped
Before: ~9/day
After: 0
linted out
— Founder approval rate
Before: 30%
After: 92%
+62pts
— The problem
What was actually broken.
A SaaS operator launching a new product needed daily voice-locked content shipping across LinkedIn, Facebook, Instagram, YouTube and TikTok — and could not afford a content team. Previous attempts with content tools resulted in obvious AI-generated posts that the founder refused to ship under his own face.
The brief was a single script that generates a full daily pack across all 5 channels, with each channel format-correct (carousel for LI, vertical for IG/TT, horizontal for YT), and zero AI-tell phrases surviving to render.
— What we built
The solution stack.
— Tools
Claude Code (orchestration)Custom 60+ AI-tell lintern8n (scheduling)html2canvas (LI carousels)FFmpeg (IG/TT vertical)
— Implementation (11 days)
- →Day 1-2: Mapped each channel's format constraints and the founder's voice samples (12 LI posts, 8 IG reels).
- →Day 3-4: Built the voice-lock prompt against the founder's actual cadence (contrarian hook → reframe → CTA).
- →Day 5-7: Wired the 5-channel parallel render — carousel HTML, vertical video, horizontal video.
- →Day 8: Linter wired pre-render — any AI-tell phrase hard-fails the build.
- →Day 9-10: n8n schedule + drip cadence across channels.
- →Day 11: Founder dry-run, ship.
— Results
What changed.
The reason founders refuse to ship AI-generated content under their face isn't capability — it's pattern recognition. They've read enough LinkedIn slop to spot 'fast-paced world' and 'unlock potential' at 30 paces, and they assume their audience can too. The linter is the trust mechanism that lets them sign off.
Channel-format parity was the second hard problem. Most content engines spit out one piece of copy and call it 'cross-platform'. Real cross-platform means a 1080×1350 carousel for LinkedIn, a 9:16 vertical for IG/TT, and a 16:9 horizontal for YT — each with the hook, the body and the CTA recomposed to fit the format's pacing. We modeled all three.
The founder approval rate jumping from 30% to 92% is the only metric we cared about. If the founder won't ship it, the rest of the system is irrelevant.
“I sign off on 9 out of 10 packs without rewriting anything. I haven't had that experience with any other content tool.”
Tools & services used
— Same builder · same playbook
Want a result like this?
Send the brief. Fixed-price scope back in 48 hours. No retainer hostage situation.
Apply for a discovery call