Article · 2026-07-20 · 6 moments

How the founder of Morning Brew built a Claude content machine that never runs out of ideas and never sounds like slop | Alex Lieberman

Listen now | 🎙 How Alex Lieberman (Morning Brew) built a Claude workflow that interviews him before drafting, codes his voice in Markdown, and runs a six-persona revision loop before posting ✦ AI generated

Read original ↗

GUEST Alex Lieberman

timeline · colored by role

Mechanism

Codifying your voice in a Markdown file makes an AI draft in your own register instead of the internet's average, and this is tied to why your employees are your most underleveraged marketing channel right now.

Alex describes codifying his personal voice and style into a Markdown file so Claude's drafts sound like him specifically rather than a generic 'internet average' tone.

transcript

Alex Lieberman: How to codify your voice in a Markdown file so an AI drafts in your register, not the internet’s average. Why your employees are your most underleveraged marketing channel right now

supports · 1

Mechanism

The blank page is the biggest friction point in content creation, an AI Oracle eliminates it, and the right first step before adding any AI is mapping your current workflow.

Alex identifies the blank page as the core bottleneck in content creation and explains that before adding any AI tooling, you first need to map your existing workflow; his Oracle step is designed to remove that blank-page friction.

transcript

Alex Lieberman: Why the blank page is the biggest friction point in content creation, and how an AI Oracle eliminates it. How to map your current workflow before you add any AI

Mechanism◆

The Content Machine has six steps live in Claude: an Oracle that scans internal systems and the internet for content spikes, an interview panel that pulls out his real ideas, voice and style files that keep drafts sounding like him, an editorial council that scores and revises posts, and a lessons loop that learns from his feedback.

Alex walks through his full 'Content Machine': an idea-spotting Oracle, an interview panel that extracts his real thinking, voice/style files, an editorial council that scores and revises drafts, and a feedback loop that keeps improving the system.

transcript

Alex Lieberman: He walks us through every step of his Content Machine live: an Oracle that scans internal systems and the internet for content spikes, an interview panel that pulls out his real ideas, voice and style files that keep drafts sounding like him, an editorial council that scores and revises posts, and a lessons loop that learns from his feedback.

explains mechanism · 1extends · 1

Claim◆

AI slop actually originates at the interview step, not the drafting step, which is why building a six-step Content Machine in Claude that goes from idea spike to publishable post starts with getting the input right.

Alex makes the counterintuitive point that AI-generated 'slop' isn't caused by the drafting step but by a weak interview/input step — if the AI never extracts your real ideas, the draft can't help but sound generic.

transcript

Alex Lieberman: How Alex built a six-step Content Machine in Claude that goes from idea spike to publishable post. Why the interview step (not the drafting step) is where AI slop actually comes from.

Example

Employees are the most underleveraged marketing channel right now, and the Tenex Creator Cup turned content creation into a team sport with a $5,000 prize pool as a way to activate employee advocacy.

Alex argues employees are an underused marketing channel and describes how Tenex turned content creation into a team competition — the Tenex Creator Cup — backed by a $5,000 prize pool to drive employee advocacy.

transcript

Alex Lieberman: Why your employees are your most underleveraged marketing channel right now. How the Tenex Creator Cup turned content creation into a team sport with a $5,000 prize pool. Employee advocacy: the Tenex Creator Cup and $5K prize pool

Claim

Distribution is becoming a durable moat, and founders and teams need to 'climb Cringe Mountain' to build it, which is part of why Alex rebuilt his content process around AI without letting it produce generic slop.

Alex Lieberman argues distribution itself is now a durable moat, and building it requires founders to push through the discomfort of public content creation ('climb Cringe Mountain'), which motivated his AI-driven content rebuild.

transcript

Alex Lieberman: Alex explains why distribution is becoming a durable moat, why founders and teams need to “climb Cringe Mountain,” and how he rebuilt his content process around AI without letting it produce generic slop.

rebuts · 1

Highlight slides

Alex Lieberman's Content Machine, Live in Claude✦ from: The Content Machine has six steps live in Claude: an Oracle that scans internal systems and the internet for content spikes, an interview panel that pulls out his real ideas, voice and style files that keep drafts sounding like him, an editorial council that scores and revises posts, and a lessons loop that learns from his feedback.Editorial Council and Feedback Loop✦ from: The Content Machine has six steps live in Claude: an Oracle that scans internal systems and the internet for content spikes, an interview panel that pulls out his real ideas, voice and style files that keep drafts sounding like him, an editorial council that scores and revises posts, and a lessons loop that learns from his feedback.Refining and Learning✦ from: The Content Machine has six steps live in Claude: an Oracle that scans internal systems and the internet for content spikes, an interview panel that pulls out his real ideas, voice and style files that keep drafts sounding like him, an editorial council that scores and revises posts, and a lessons loop that learns from his feedback.AI Slop Starts at the Interview, Not the Draft✦ from: AI slop actually originates at the interview step, not the drafting step, which is why building a six-step Content Machine in Claude that goes from idea spike to publishable post starts with getting the input right.The Six-Step Content Machine (in Claude)✦ from: AI slop actually originates at the interview step, not the drafting step, which is why building a six-step Content Machine in Claude that goes from idea spike to publishable post starts with getting the input right.

Related episodes

Video · My First Million

Codifying your voice in a Markdown file makes an AI draft in your own register instead of the internet's average, and this is tied to why your employees are your most underleveraged marketing channel right now.

The blank page is the biggest friction point in content creation, an AI Oracle eliminates it, and the right first step before adding any AI is mapping your current workflow.

AI slop actually originates at the interview step, not the drafting step, which is why building a six-step Content Machine in Claude that goes from idea spike to publishable post starts with getting the input right.

Employees are the most underleveraged marketing channel right now, and the Tenex Creator Cup turned content creation into a team sport with a $5,000 prize pool as a way to activate employee advocacy.

Distribution is becoming a durable moat, and founders and teams need to 'climb Cringe Mountain' to build it, which is part of why Alex rebuilt his content process around AI without letting it produce generic slop.

Killer marketing tricks Amex, Apple and Domino's use on you - Rory Sutherland

Killer marketing secrets that always work (ft. Ogilvy Vice Chairman)

Claude make Fable 5 permanent