Ship it — live demos, peer voting, and the culmination of your agentic AI journey
By the end of this session, you'll have both conceptual understanding and hands-on experience with the foundations of AI.
Present your agent to the class with a live demo — not slides, not screenshots. Run it in real-time and show it reasoning, using tools, and producing results.
Answer Q&A about your architecture, tool choices, guardrails, and business case. Explain trade-offs and what you'd do differently with more time.
Vote on peer projects across four dimensions: Does it work? Is the problem real? Would you trust it? Compelling demo?
From 'what is AI?' to 'I built, evaluated, and pitched an agent product.' Articulate what you learned and how to talk about these skills.
A carefully crafted progression from concepts to hands-on building.
Format, voting criteria, and order of presentations.
Team presentations — live agent demos with Q&A.
Catch your breath. Almost done.
Peer voting results and awards.
From 'what is AI?' to 'I built an agent product.' What this means for your career.
Everything you need to explore, experiment, and build.
Rate each team's demo on four criteria: Does it work? Is the problem real? Would you trust it? Compelling demo?
Open formEvery session's materials, notebooks, and resources — your complete reference for building AI agents.
Browse all sessionsThe guide that accompanied you all semester. Revisit as you continue building agents after this course.
Read guideThree interactive challenges to build your intuition.
Each team presents their final project:
What problem does your agent solve? For whom? Why does it matter?
Run your agent LIVE. Show it reasoning, calling tools, and producing results. Show at least one edge case or failure recovery.
Show your evaluation metrics: accuracy, consistency, safety scores. What's your agent's batting average?
Who needs this? What's the ROI? What are the risks? Would you deploy it in production?
Answer questions from peers and professor. Defend your architecture, tool choices, and guardrails.
Look back on the journey and forward to your career:
What was the most surprising thing you learned about AI agents this semester?
If you had another semester, what agent would you build? What tools would you add?
How will you use agent-building skills in your career? What would you put on your resume?
You shipped a product. Not a homework assignment. Not a tutorial. A real agent that solves a real problem, with tests proving it works and a business case explaining why it matters.
The code was always simple. The agent loop is ~50 lines. ReAct is a system prompt. RAG is search + context. The hard part was always the THINKING — choosing the right pattern for the right problem.
You're entering at the right time. AI agents are going from research demos to enterprise products right now. The skills you built this semester are exactly what companies are hiring for.
Keep building. The frameworks will change. The models will improve. But the patterns — ReAct, RAG, multi-agent, guardrails, evaluation — those are durable. You now speak the language.
Complete these tasks before our next class to prepare for prompt engineering.