Skip to content
PMInterviewKit
In this module9 lessons
  1. What AI product rounds testFree
  2. How large language models work, for PMs
  3. Designing an AI feature
  4. Evals and quality
  5. Metrics for AI products
  6. Risk, safety and trust
  7. Cost and latency trade-offs
  8. Worked answers
  9. Common traps and how to recover

Designing an AI feature

Run the Job, Fit and Failure gates on any AI design prompt, decide when not to use AI, and design review steps, fallbacks and uncertainty.

7 min readFull kit

Most AI design answers go wrong in the same place. The candidate picks a feature, describes the happy path in loving detail and runs out of time before saying what happens when the model is wrong. This lesson teaches the first three of the Five Gates (Job, Fit and Failure), which is where AI design answers are won.

We will use one running example throughout: Harbor, a fictional app that small landlords use to manage rental properties. The prompt: "Use AI to help Harbor landlords handle tenant messages."

Checking your access…

Practice this lesson

← PreviousHow large language models work, for PMsNext →Evals and quality