Loading…

Session guide

Ask about the workshop or how the method works. Keep exercise decisions with your group.

Chat messages are processed by Pickaxe. Don’t enter names or workplace records. Your group’s work is not sent to the guide.

Quick answers to common questions

Starter questions

What should my first small test look like?

Pick one recurring task with a clear result. Your first action can be confirming access to approved tools and data, or preparing a synthetic sample. Plan two or three comparable cases.

What must my reviewer check?

The claim whose error would cost the most: open the exact source, decide whether it supports the next step, and record advance, revise or hold; add a reason if useful.

How do I measure human effort?

Net human time = comparable manual time − (briefing + active work + review + correction + handoff). Record waiting separately, and pair the number with a quality or rework check.

When should I change the method?

When eligibility rules, source structure, approved tools, authority or the output purpose change. Re-test on an unknown-evidence case and an explicit-mismatch case first.

Loading the guide…

Material preview

The full numbered text available to the AI. Opening this preview does not change the materials or your instructions.