Iterate: from draft to great output

Iterate: from draft to great output

Why it matters: Iteration turns “okay” outputs into high-quality deliverables with less effort than starting over.

Goals

• Use a 4-step iterate loop.
• Generate critiques that are actionable.
• Stop when the output passes criteria.

Key definitions

• Draft: A first version meant to be improved, not shipped.
• Critique: A structured quality review against criteria (not vibes).
• Revision: Changes that address the critique with minimal extra complexity.
• Verification: Checks for correctness, consistency, and compliance.

Workflow (step-by-step)

1. Draft: ask for a first version in the target format.
2. Critique: ask for issues vs your checklist + what’s missing.
3. Revise: ask for an updated version that fixes the critique.
4. Verify: ask for a self-check against the checklist and any uncertain claims.

Example (good vs bad)

✅ Good: You iterate until the output passes a checklist and has no unsupported claims.

❌ Bad: You keep rewriting prompts randomly without a consistent rubric or stopping rule.

Checklist

• I have a written checklist.
• Critique is tied to checklist items.
• Revisions address specific failures.
• I stop when criteria is met.

Metrics / criteria

• Cycles: <= 3 iteration cycles for most tasks.
• Measurable: You can say whether the output is correct/complete (pass/fail or a score).
• Effort: You can produce the result in 15–30 minutes using the workflow.
• Clarity: Another person can run your prompt and get a similar outcome.

Common mistakes + fixes

• Pitfall: Asking “make it better” with no rubric. Fix: Provide criteria and ask for a point-by-point critique.
• Pitfall: Infinite iteration. Fix: Define a stopping rule (score threshold or checklist pass).
• Pitfall: Vague instructions. Fix: Add constraints, examples, and a success checklist.
• Pitfall: One-shot prompting. Fix: Iterate: draft → critique → revise → verify.
• Pitfall: No validation. Fix: Add QA steps, tests, and explicit “what counts as done”.

Student tasks

1. Apply the workflow to one real work task related to iterate: from draft to great output.
2. Create a small artifact: prompt, checklist, rubric, table, draft, or decision note.
3. Run one quality check and record what changed after the check.
4. Write a 3-sentence reflection: what worked, what failed, and what you will reuse.

Useful external sources

• OpenAI prompt engineering guide: https://platform.openai.com/docs/guides/prompt-engineering — Prompt structure, iteration, constraints, and evaluation habits.
• NIST AI Risk Management Framework: https://www.nist.gov/itl/ai-risk-management-framework — Risk framing, governance language, and verification mindset.
• OWASP Top 10 for Large Language Model Applications: https://owasp.org/www-project-top-10-for-large-language-model-applications/ — Practical AI safety risks such as prompt injection, leakage, and insecure output handling.
• Google People + AI Guidebook: https://pair.withgoogle.com/guidebook/ — Human-centered AI design, user control, feedback, and failure modes.

Bibliography

• OpenAI prompt engineering guide. https://platform.openai.com/docs/guides/prompt-engineering
• NIST AI Risk Management Framework. https://www.nist.gov/itl/ai-risk-management-framework
• OWASP Top 10 for Large Language Model Applications. https://owasp.org/www-project-top-10-for-large-language-model-applications/
• Google People + AI Guidebook. https://pair.withgoogle.com/guidebook/