PMs Who Build

Y / Build

The Weekly Build

One brief a week. About 90 minutes. Any tool. Build it, share what happened, and see what other PMs made. The format comes from a simple idea: the hypothesis is the product, and the code is just the test.

Week 01Live now
Week 02Next Monday
Week 03Coming up

This week's brief

Week 01 · L3 · 90 min

Turn a pile of customer feedback into a triage tool

The bet
PMs lose hours each week sorting raw feedback by hand. A small tool that groups feedback by theme and flags urgent items would cut that to minutes, and you'd trust it enough to use it.
Ask for this outcome
Paste in 50 to 200 rows of real feedback (support tickets, survey answers, app reviews). Get back themes with counts, the three most urgent items with the reason they're urgent, and every item tagged so you can check the work.
Catch the failure
Watch for invented themes with no rows behind them, and for one angry customer being counted as a trend. Ask it to show the row behind every claim.
Add the guardrail
Make the tool cite row numbers for every theme and refuse to call something urgent without a quoted line. Strip names and emails before you paste anything in.
Watch real use
Run it on this week's feedback and compare it with how you'd have sorted it. Where did it beat you? Where would you not trust it?
Tools
Any: Claude, ChatGPT, Lovable, v0, Cursor, Replit
Share
Share a screenshot (no customer data), what you built it with, and one thing that surprised you.

How every brief works

  1. Write the bet. Who has the problem, what you believe, and what would prove you wrong.
  2. Ask for an outcome, not an app. Prompt for the result you want, in the user's words.
  3. Catch the failure. AI features don't crash. They get confidently wrong. Find where.
  4. Add the guardrail. Fix the failure before anyone else uses it.
  5. Watch real use. Put it in front of one real person and note what they actually do.

The Builder PM Ladder

Five rungs, from first prompt to running agents. Find yours, then pick briefs one rung up.

L1

Prompt

Use AI chat to draft, summarize and analyze your own product work.

Proof: A PRD, research synthesis or spec you drafted with AI and then corrected.

L2

Prototype

Turn an idea into a clickable prototype from a prompt.

Proof: A working prototype you put in front of a stakeholder or user.

L3

Data

Build a tool that reads real data: a CSV, a spreadsheet, an API.

Proof: Something that answers a question with your team's actual numbers.

L4

Ship

Deploy something other people use. Read and review the code AI wrote.

Proof: A URL your team uses, in version control, that you can change safely.

L5

Systems

Run agents that do real work, with evals, model choice and cost per call in mind.

Proof: An agent or workflow with written evals and a known cost.

The gallery opens with the first week's builds. Sign up and yours could be the first one here.