APPENDIX H: RESUME PORTFOLIO BUILDING CAPSTONE PROJECTS

WHAT YOU WILL BE ABLE TO CLAIM AFTERWARDS You built a supervisor and worker architecture in Python with per-ticket budgets, retries with exponential backoff, a review step, human escalation, an audit log, and a kill switch. You wrote automated tests that prove retries, redrafts, budget escalation, and the kill switch all work. You wrote an incident runbook and a deployment plan for the system. You can discuss reliability, cost control, and operations as part of agent design. Do not claim that you deployed or operated a production system unless you really did, because this project is a design and a tested prototype running on invented tickets. SKILLS PRACTICED Budgets and resource limits, covered in Chapter 25. Why and how to use more than one agent, covered in Chapters 42 and 43. Coordination and communication between agents, covered in Chapters 44 and 45. Reliability and error recovery, covered in Chapter 56. Moving from prototype to production, covered in Chapter 57. Deployment architectures, covered in Chapter 58. Agent operations, covered in Chapter 59. Incident response for agents, covered in Chapter 60. Responsible AI and risk, covered in Chapter 41.