Lab 5 · AI Prompt Writing Lab: From Weak to Strong Prompts

🎯 Objective

Learn prompt engineering by controlled experiment: change one prompt systematically and measure how the answer improves.

🧩 Skills You Will Practice

• Prompt formula (Role
• Task
• Context
• Format
• Examples)
• Iteration
• Output evaluation
• Fact verification

🧰 What You Need

• One AI chatbot (free version)
• A notes document
• One real study topic from your current semester
• The Prompt Engineering lesson open for reference

📋 Step-by-Step Tasks

1. Write 5 weak prompts you would naturally type, for example Explain loops or Make notes on databases. Save the exact answers you get — this is your baseline.
2. For prompt 1, add only a Role and Task. For prompt 2, add Context (your semester, your exam, what you already know). For prompt 3, add a Format (table, 5 bullets, word limit).
3. For prompt 4, add an Example of the style you want. For prompt 5, use all five parts together and add If you are unsure, say so.
4. Run each improved prompt, save the new answer next to the old one, and score both out of 5 for accuracy, clarity, fit to your level and usefulness. Be strict — the point is honest comparison.
5. Pick your hardest topic and iterate three times: Make it simpler, Now give a harder example, Now quiz me with 3 questions without answers. Solve the quiz on paper before checking.
6. Verify two factual claims from the AI answers against your book or class notes and mark each claim Correct / Partly correct / Wrong.
7. Save your 5 improved prompts as reusable templates in your notes with blanks for topic and level.

✅ Expected Result / Deliverable

A before/after prompt portfolio: 5 weak prompts, 5 engineered prompts, scored answers, 2 verified claims and 5 reusable templates you can use all semester.

⚠️ Common Mistakes

• Changing everything at once so you learn nothing about what helped
• Accepting a fluent answer without verifying facts
• Writing prompts so long the actual task gets buried