13. Advanced Prompt Patterns
Use role, task, constraints, examples, rubrics, and iteration patterns like a specialist.
By Jacques Botte, founder of Toptronic®. Last updated 12 September 2026.
The lesson
Advanced prompting is not about magic words. It is about controlling ambiguity. A strong prompt states the role, task, facts, constraints, examples, scoring rubric, and the exact output shape.
Useful patterns include critique-then-rewrite, compare-options, assumptions-first, test-case generation, risk table, decision memo, and stepwise refinement.
TPEE supports these patterns by letting you save reusable templates instead of reinventing a good instruction every time.
Check yourself
Question 1: What is the purpose of critique-then-rewrite?
- Make output random
- Avoid instructions
- Skip review
- Find weaknesses before producing the improved answer — correct
Answer: Find weaknesses before producing the improved answer
Critique exposes defects before the final rewrite.
Question 2: Why include a rubric in a prompt?
- It forces internet access
- It removes examples
- It tells the AI how success will be judged — correct
- It hides the goal
Answer: It tells the AI how success will be judged
A rubric turns quality into visible criteria.
Question 3: Which is an advanced prompt pattern?
- One word only for all tasks
- Assumptions-first — correct
- No context ever
- Secret-key paste
Answer: Assumptions-first
Asking for assumptions reveals hidden uncertainty.
← Previous lesson · All 83 lessons · Next lesson →
The full course — 83 lessons and 249 quiz questions — ships inside the app. Get TPEE to study it offline.