Free lesson · Level 2 — Ask
Effective Prompting and Critical Thinking
Frame the problem, supply context and constraints, and ask for output you can check.
01Stop using AI like a search box
A one-line question gets a generic answer. Facility problems need context: the asset, the data, the constraint, and the decision you're trying to make.
02The FAI request structure
Role → Context → Data → Task → Constraints → Output format → Verification. You don't need every part every time, but each one you add removes a guess.
ROLE: You are assisting a facility manager. CONTEXT: 3-building campus, 30 major assets. DATA: Attached CSV of 2025 work orders. TASK: Identify the 3 assets with the most corrective work. CONSTRAINTS: Remove duplicate IDs first. Show counts. OUTPUT: Table with asset, count, and % of corrective work. VERIFY: List any data problems you found and how you handled them.
03Ask for the evidence
Ask the AI to show its working, state assumptions, and flag what it couldn't verify. Then check the numbers yourself on a sample.
Check your understanding
Which addition most improves a request for maintenance analysis?