Week 16: Final Presentations and Course Wrap-up

Phase 5Week 16Final PresentationsLecture: 2026-06-16

Final Presentation Schedule

Individual projects. Each student: 15 min talk + 5 min Q&A = 20 min. Slot count adjusts to the number of enrolled students.

Example schedule (slot count adjusts to the number of enrolled students):

Time	Slot	Notes
09:00–09:20	#1	15-min talk + 5-min Q&A
09:20–09:40	#2
09:40–10:00	#3
10:00–10:15	Break
10:15–10:35	#4
10:35–10:55	#5
…	…	Extend for additional students
Last 30 min	Write peer evaluations
Last 10 min	Course wrap-up & grade announcement

Presentation order: Reverse of the Week 8 order (whoever presented first then presents last).

Peer Evaluation Criteria

Each student’s presentation is evaluated on the following 5 items (20 points each, 100 points total):

1. Technical Completeness — Does the agent pipeline actually work?
2. Problem-Solving Ability — Is the chosen problem appropriate and is the solution effective?
3. Clarity of Presentation — Was the complex system explained in an accessible way?
4. Application of Ralph Loop Philosophy — Were harness engineering principles properly implemented?
5. Learning Reflection — Were failures and lessons shared honestly?

The Future of AI Systems

Near-term (2026–2027)

• HOTL becomes the standard development methodology
• Full enforcement of the EU AI Act makes Governance-as-Code mandatory
• Open-source models reach performance near commercial APIs

Mid-term (2027–2030)

• Fully autonomous SDLC — humans only provide requirements
• Meta-agentic systems where agents design other agents
• AI Engineer = integrated role of system architect + supervisor

A Note as This Course Ends

You are no longer just developers who write code — you are systems engineers who direct AI agents. Ralph Loop, HOTL governance, multi-agent SDLC — these will become the standard of the future software industry. Use this course as a starting point. Keep learning. Keep building.

Career Preparation Guide

How to represent the skills built in this course on your resume:

## Project Experience
- **Ralphthon Capstone**: Multi-agent autonomous SDLC design and implementation
  - Tech: Ralph Loop, HOTL, vLLM, DeepSeek-Coder-V2, MCP
  - Outcome: 90% automation of code generation → testing → deployment pipeline
  - Infrastructure: NVIDIA DGX H100 (MIG), Kubernetes

## Tech Stack
- AI Systems: Agentic workflows, harness engineering, LLM operations
- MLOps: vLLM, OpenTelemetry, LLM-as-Judge
- Infrastructure: DGX H100, MIG, Docker, Kubernetes

Congratulations!

Thank you to all the students who studied with us for 16 weeks. We hope everything you built in this course becomes a major asset in your career.

Questions or feedback: yj.lee@chu.ac.kr or GitHub Issue