Library › Operations AI › Operations and supply chain

Report a cooler temperature excursion from the log and route it to QA

  • For Operations and supply chain
  • You practice Catching mistakes and made-up facts
  • Time about 5 minutes
  • Level Intermediate

About this exercise

Write a prompt for plain text with exactly four lines: "Out of range:", "Duration:", "Peak:" and "QA review:". Out of range lists the readings above 40 F. Duration runs from the first out-of-range reading to the first reading back in range, shown as a subtraction. QA review says whether the policy requires it and that QA decides what happens to the product. Do not say whether the product is safe. Under 90 words.

Common mistakes when you ask AI for this

  • Measuring duration wrongly. The AI counts first to last out-of-range reading (30 minutes). Try: Say: "Duration: first out-of-range reading to the first in-range reading."
  • Judging safety. The AI says the product is likely fine or must be discarded. Try: Say: "Do not say the product is safe or unsafe; QA decides."
  • Stating the door as the cause. The AI repeats the dock lead's guess. Try: Say: "Do not state a cause."
  • Wrong lines. The AI writes a paragraph. Try: Quote the four labels.
  • Markdown extras and long output. AI adds bold, headings and bullets by default, which show up as stray symbols where this is pasted. Try: Say "plain text, no bold, no headings" and cap each line, for example "each line under 15 words".

What a strong prompt includes

  • Quotes the four labels
  • Defines the duration
  • Applies the 30-minute policy
  • Leaves safety to QA
  • Asks for plain text

Practice this now ↑

Related exercises