AI PM Storytelling: How to Talk About Your AI Work in Public
TL;DR
Storytelling is the highest-leverage skill an aspiring AI PM has — and the most under-practiced. The candidates landing the best roles aren't always the most technical; they're the most articulate about their technical work in public. This guide covers the four story formats that build AI PM reputation, the hooks that earn attention, and the cadence that converts public posts into hiring conversations.
Why Public Storytelling Matters in AI PM
Hiring managers Google candidates. Recruiters scan LinkedIn for proof of thinking. Investors search Twitter before meetings. The aspiring AI PM with 12 months of consistent public storytelling is in 100x more conversations than the equivalent candidate hidden in private. Storytelling isn't self-promotion — it's discoverability.
Format 1: The shipped feature breakdown
What you built, why, what worked, what didn't. The cornerstone format. Strongest signal of competence.
Format 2: The reasoned opinion
"Here's what I think about X" with specific reasoning. Builds the perception of an active thinker.
Format 3: The failure postmortem
What went wrong, what you learned. Counterintuitively the highest-engagement format. Builds enormous trust.
Format 4: The synthesis post
"I read 10 papers; here's the pattern." Shows curation skill and saves your audience time. Compounds attention.
The Hook That Gets You Read
The first sentence determines whether anyone reads the rest. Most aspiring AI PMs open with throat-clearing. The candidates who get attention open with a specific, contrarian, or surprising claim that earns the next 30 seconds.
Specific number
"Cut our model spend by 67% in 4 weeks." Numbers anchor; vague claims slide past.
Contrarian observation
"Most AI PMs over-invest in eval and under-invest in user trust UI." Take a position; defend it.
Confession
"I shipped an AI feature with no eval set. Here's what happened." Vulnerability earns reading time.
Counter-narrative
"Everyone says fine-tuning is rarely worth it. We fine-tuned. Here's when it actually was."
Tight scoped story
"Friday at 4pm: a user reports the chatbot is making things up. By 6pm we found it. Here's how."
The Story Arc That Lands Every Time
Every great AI PM post follows the same structure: situation, complication, decision, outcome, lesson. Five beats, in order. When you learn to write to these beats, you stop writing posts that ramble.
Situation (1-2 sentences)
Set the scene. Where you were, what you were building, who the user was.
Complication (1-2 sentences)
What went wrong, what you didn't expect, what made the decision hard.
Decision (2-4 sentences)
The choice you made, the reasoning, what you almost chose instead. The intellectual core.
Outcome (1-2 sentences)
What happened. Specific numbers if possible.
Lesson (1 sentence)
The portable insight others can use. The reason you wrote the post.
Build a Storytelling Practice in the Masterclass
The AI PM Masterclass includes weekly writing prompts and instructor feedback on your public posts — the discipline that turns LinkedIn into a hiring funnel.
The Cadence That Compounds
Twice a week, short
200-400 words. Anchored in something specific you did or thought that week. Most engagement happens here.
Once a month, long
1,500-2,500 word essay. One synthesized theme. Becomes the post you point to in interviews and applications.
Daily comments on others
Reply more than you post. Working AI PMs notice consistent thoughtful commenters before they notice posters.
One conference talk per year
Even if it's a 10-person meetup. Forces deeper synthesis than any post and produces an artifact (slides, video) you can share.
Mistakes That Sink Public Storytelling
Generic AI commentary
"AI is changing everything" reads as filler. Specific projects, specific numbers, specific decisions or nothing.
Performative humility theater
"Just a humble student of AI..." signals insecurity. Confident reasoning earns respect; theatrical humility doesn't.
Ghostwriter-style polish
Posts that sound machine-generated get scrolled past. Voice matters more than perfect grammar. Lean into specifics, not stylistic flourishes.
No consistency
Five posts in week 1, zero for two months, two more posts. Compounding only works with sustained cadence.
Engagement-bait formats
"What's your favorite AI tool? 👇" reads as low-effort. Tradeoffs and reasoning beat polls every time.