Capstone: your end-to-end AI workflow

Capstone: your end-to-end AI workflow

Why it matters: A capstone forces an end-to-end workflow and exposes where your system breaks.

Goals

• Build an end-to-end workflow.
• Add QA and guardrails.
• Ship a final output you can reuse.

Key definitions

• End-to-end: From raw input to shipped output with checks and documentation.
• System: Templates + workflow + QA + review cadence.
• Evidence: Proof that the output meets criteria (tests, sources, checks).

Workflow (step-by-step)

1. Pick one real deliverable (brief, FAQ, outreach sequence, summary).
2. Write the workflow: inputs → draft → critique → revise → QA → ship.
3. Add a rubric and red-flag rules.
4. Run the workflow and record what failed.
5. Update templates so next run is faster and safer.

Example (good vs bad)

✅ Good: A capstone deliverable with a reusable workflow and QA evidence.

❌ Bad: A capstone output with no QA, no rubric, and no plan to reuse it.

Checklist

• Workflow documented.
• Rubric exists.
• QA performed.
• Templates updated.
• Next review scheduled.

Metrics / criteria

• Reusability: you can run the workflow again next week.
• 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: Treating capstone as one-off. Fix: Capture templates and a review cadence so it becomes a system.
• 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 capstone: your end-to-end ai workflow.
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/