You understand the difference between code-centered and spec-centered work.

You know the fundamentals of the AI Unified Process (AIUP).

You have experienced firsthand how Claude Code generates code and tests from specifications.

You know how AI supports implementation – while the specification remains the source of truth.

Spec-Driven Development – Why?

• Problems of code-centered work: Specifications become outdated, documentation lags, tests are missing.
• Advantage of spec-centered approach: A Markdown specification as the single source of truth.
• Derived from the specification: Use Cases → Entity Models → System Use Cases → Implementation → Tests.

AI Unified Process (AIUP)

• Inception: Goals, stakeholders, initial requirements.
• Elaboration: Use cases, glossary, entity models, acceptance tests.
• Construction: Claude Code generates code + tests from the spec; developers review & correct.
• Transition: Review, acceptance, deployment, feedback for the next iteration.

Claude Code in Action

• No “vibe-coding,” only Spec-Driven Development.
• Example workflow:
  • Describe requirements → generate use cases.
  • Claude Code generates code & tests.
  • Review results, correct, iterate.

Hands-On: Hotel Reservation System

• Define 1–2 use cases as a specification.
• Claude Code generates domain classes, REST API, repository, and tests from them.
• Goal: By the end of the workshop, a small app is running – fully developed from a specification.

Java Software Developers and Architects

Knowledge of Java Programming