APPENDIX H: RESUME PORTFOLIO BUILDING CAPSTONE PROJECTS

ARTIFICIAL INTELLIGENCE ASSISTANT PROMPTS These two prompts reflect operations work a harness engineer does around a live agent system. Replace each bracketed placeholder with your own material before use, and remove private data first. [START OF PROMPT] I maintain a multi-agent workflow with a supervisor and these workers: [LIST WORKERS AND THEIR JOBS]. Its protections are: [LIST BUDGETS, RETRIES, KILL SWITCH, HUMAN ESCALATION RULES]. Act as a site reliability reviewer. Walk through ten failure scenarios, including a worker returning wrong answers confidently, a tool being slow, a retry storm, a budget set too low, and the kill switch being forgotten. For each scenario, say how my current protections respond, what gap remains, and what I should add to my runbook. Output format: a numbered list with the fields Scenario, Current Protection, Remaining Gap, and Runbook Addition. [END OF PROMPT] [START OF PROMPT] Here is the draft of my incident runbook for an agent system: [PASTE THE RUNBOOK TEXT WITHOUT SECRETS]. Review it as if you were a tired on-call engineer reading it at three in the morning. Point out every step that is vague, missing, or in the wrong order, rewrite the first five minutes section so each action is a short imperative sentence, and suggest three practice drills I could run to test it. Output format: Problems Found, Rewritten First Five Minutes, Three Drills. [END OF PROMPT] APPENDIX H PORTFOLIO ASSEMBLY GUIDE