AI coding agents can now plan, edit, test, and review software inside a development environment. This workshop teaches you how to direct that work and how to check the result.
The workshop is for experienced software developers. You will build three projects with Claude Code: a CRUD app, a HADR agent, and a team AI app. You will also evaluate model behaviour, cost, and quality.
Course path
- SetupForm a team, prepare the tools, and understand the course context
- Project 1: CRUD appTurn a product brief into a small working application
- Project 2: HADR agentPlay the dispatch game, then build and evaluate an agent
- Project 3: AI appApply the course techniques in one team project
- Presentations and conclusionExplore Claude Code on the web, present the AI app, and plan what comes next
The game reference contains the HADR rules, maps, MCP tools, and API details. Use it when an activity links to it; it is not a separate project.
About your instructor
Hi, I am YJ.
- Former teacher and civil servant; current company cofounder, teacher, programmer, and designer.
- Tinkercademy teaches programming, digital making, and AI.
- Cofounder of the Agentic Builders Collective, an AI builders interest group.
yjsoonon X, LinkedIn, and GitHub.
Facilitators will introduce themselves at the start of the workshop.
About the class
- Two full days, 9:00 to 17:00.
- Breaks are planned for mid-morning and mid-afternoon, with lunch around midday.
- Teams of 3-4.
- Each participant builds a CRUD app and a HADR agent in separate repositories.
- Teammates review the HADR work and share evaluation tasks.
- Each team builds one AI app and presentation together.
- Setup happens in class. The project headings show the learning sequence; they are not a day-by-day timetable.
Shared workspace
Use the course Padlet at padlet.tk.sg/hadr2608a to share useful articles, tools, failures, discoveries, questions, and vocabulary. Keep it open during class and never post secrets or operational information.