Interview Me

What people ask for and what they actually want are different things. This skill closes the gap before it costs anything.

When to Use

• The ask is missing at least one of: who, why, what success looks like, or the binding constraint
• The request is conventional rather than specific ("build me X", "make it faster")
• You're tempted to start with assumptions you haven't surfaced
• The user explicitly invokes: "interview me", "grill me", "stress-test my thinking"

The Process

Step 1: Hypothesize, with a confidence number

Before asking anything, write down your current best read plus an honest confidence number (0–100%):

HYPOTHESIS: You want a way to answer "how are we doing?" in standup, and "dashboard" was the convention that came to mind.
CONFIDENCE: ~30% — missing: who it's for, what "metrics" means, and what success looks like

Step 2: Ask one question at a time, each with a guess attached

Q: <one focused question>
GUESS: <your hypothesis for the answer, with reasoning>

Why one at a time, not a batch:

• The user can't react to your hypotheses if you bury them in a list
• The third question often depends on the answer to the first
• The user's energy for thinking carefully is finite

Why attach a guess:

• The user reacts faster to a wrong guess than they generate an answer from scratch
• It commits you to a hypothesis you can be visibly wrong about

Step 3: Listen for "want vs. should want"

Watch for:

• Answers that pattern-match best-practice talk without specifics
• Answers that defer to convention ("the standard approach")
• Buzzwords as goals — "modern", "scalable", "robust"

When you hear these, ask:

"If you didn't have to justify this to anyone, what would you actually want?"

Step 4: Restate intent in the user's own words

When confidence is high, write back what you think the user wants:

Here's what I now think you want:

- Outcome:      <one line>
- User:         <one line — who benefits>
- Why now:      <one line — what changed>
- Success:      <one line — how we know it worked>
- Constraint:   <one line — the binding limit>
- Out of scope: <one line — what we're explicitly not doing>

Yes / no / refine?

Step 5: Confirm — explicit yes, not "whatever you think"

The gate is an explicit "yes." These are not yes:

• "Whatever you think is best." → Re-ask with two concrete options
• "Sounds good." → Ask: "Anything you'd refine?"
• "Sure, let's go." → Often a polite exit, not an endorsement

The 95% Confidence Stop

You're done when you can answer yes to this:

Can I predict the user's reaction to the next three questions I would ask?

If yes, you have shared understanding. Stop interviewing and produce the restate.

Example

User: build me a dashboard for our metrics

Agent:
HYPOTHESIS: You want a way to answer "how are we doing?" in standup.
CONFIDENCE: ~30% — missing: who it's for, what "metrics" are, success criteria

Q:     When you say "how are we doing?", who's asking — you alone, the team, or up the chain?
GUESS: engineering team in standup, because "we" usually scopes that way

User: It's actually for me. I keep losing track of which experiments are running.

Agent:
Updated read: not a team dashboard. It's a personal experiment tracker. Confidence: ~60%

Q:     Is the gap that you don't know which experiments exist, or that you can't see their results in one place?
GUESS: the second one. Results live in five different tools.

User: First one actually. I literally don't have a list.

Agent: Two questions in, the actual ask isn't "a dashboard." It's "a list."

Verification

After applying interview-me:

• A hypothesis with confidence number was stated in the first turn
• Questions were asked one at a time with guesses attached
• A concrete restate was written back to the user
• The user confirmed with an explicit yes
• The agent could predict reactions to the next three questions at the stop point