Get what's in my best person's head into something others can use

We have someone who is very good at [THE THING]. Nobody else can do it to the same standard, and none of it is written down. I want to interview them properly, and I want you to run the interview.

Who they are: [ROLE, HOW LONG THEY'VE DONE IT]
What they're good at: [SPECIFICALLY]
What goes wrong when someone else tries: [IF YOU KNOW]

Interview them through me. ONE question at a time, wait for the answer, then follow up before moving on. About ten questions.

Ask the questions that surface tacit knowledge — the things experts don't mention because they don't know they know them:
- Walk me through the last real one you did, start to finish
- Where did you stop and think, even briefly?
- What did you check that nobody told you to check?
- What did you decide not to do, and why?
- What would a competent newcomer have got wrong there?
- What tells you early that this one is going to be difficult?
- When you look at a finished piece of this work, what makes you wince?
- What do you do when the usual approach doesn't fit?

Follow up on anything vague. "You just get a feel for it" is the beginning of an answer, not the end — ask what specifically they noticed the last time they got that feeling.

Then produce:
1. THE PROCEDURE — the steps someone else could follow
2. THE JUDGEMENT CALLS — where the procedure runs out and what to weigh, framed as questions rather than rules
3. THE TELLS — the early warning signs they described, in their words
4. THE MISTAKES — what newcomers get wrong, and what to check to catch each one
5. WHAT STILL CAN'T BE WRITTEN DOWN, and how someone would build it

Be honest in section 5. Pretending everything transfers is why documentation gets ignored by the people it was written for.