AI for PMs
Building with AI, judging its output, and the PM skills the shift actually rewards.
The supply of defensible improvements is now infinite. Your capacity to judge is not.
AI Moved the Bottleneck, It Didn't Remove It
When optimization gets frictionless, the original problem goes quiet, and most teams never notice it left.
ReadAbsence of instruction is not a constraint. Silence reads as permission.
Negative Prompts Are a PM Skill
The hardest part of a good prompt is the same hard part of a good spec: naming what you will not allow.
ReadThe evals passed. The product failed. These are not the same thing.
AI Evals Are a Product Decision, Not an Engineering One
How to design quality rubrics and golden datasets that actually measure whether your AI feature does the right thing, n…
ReadA thumbs-down button with no path to model change is a dashboard nobody reads.
The AI Flywheel, Why Most Teams Break It Before It Starts
A closed-loop feedback system turns user corrections into continuous model improvement. Most teams ship a thumbs-down b…
ReadA system that is right 80% of the time can be wrong where it matters most.
The AI Product Requirements Document, Writing Specs for Systems That Might Be Wrong
How to specify AI-powered features when the output is probabilistic, not deterministic, defining acceptable variance, s…
ReadLegal said no to the version we built without asking them.
AI in Regulated Industries, The Compliance Layer Is a Product Decision
Most AI features killed by legal were not killed by compliance, they were killed by product teams that treated complian…
ReadThe signal that changes your product almost always comes from something that surprised you.
AI as a Research Partner, Accelerating Discovery Without Outsourcing the Judgment
AI can make user research faster, but used incorrectly it removes the signal that changes product direction. This artic…
ReadStreaming can make a model look less capable than it actually is.
The Latency Problem Is a Design Problem
How to design around large language model latency so users do not experience wait time as a product failure, and why st…
ReadAsking a model for weaknesses is asking a mirror whether your outfit looks good.
Using an LLM as a Sparring Partner, Without Letting It Agree With You
Most PMs use LLMs to validate their thinking rather than challenge it. This article shows how to configure a sparring p…
ReadThe model did exactly what it was built to do. The onboarding lied.
The Onboarding Honesty Problem, Setting Expectations for an AI That Will Fail
When onboarding copy says 'AI-powered' without specifics, users fill in the gap with their best-case scenario. This art…
ReadIn an LLM feature, an unspecified behavior produces a confident, fluent, wrong answer.
Reading the Trace, How to Diagnose an AI Failure After It Happens
When an AI agent produces a bad output, 'the model hallucinated' is almost never the complete explanation. This article…
ReadIf your AI feature shipped without these elements, it was released, not launched.
Rewriting Done, Completion Criteria When the Output Is Probabilistic
Traditional Definition of Done breaks for AI features because quality assurance can verify the feature ran, not that it…
ReadThe most dangerous AI output is the one that is wrong in a way that sounds right.
When the AI Sounds Confident, That Is the Warning Sign
Large language models are trained to produce fluent, complete responses, not accurate ones. When your product surfaces…
ReadAI synthesis makes you faster at confirming the insights you already had.
When to Trust AI in Your PM Workflow (and When the Trust Is a Shortcut)
A hard look at which PM tasks AI genuinely accelerates and which ones it makes look done without actually doing them. T…
Read