From Requirements to Code: How AI is Reshaping the SDLC with Simon Martinelli

In this episode, I sit down with Simon Martinelli, a Swiss Java Champion and creator of the AI Unified Process, to explore how AI is fundamentally reshaping the software development lifecycle. Simon shares his journey from "vibe coding" to developing a structured, spec-driven approach that puts requirements at the center of development. We discuss why system use cases work better than user stories for AI collaboration, how self-contained systems architecture reduces complexity, and why specifications—not code—should be your source of truth. What You'll Learn: ✅ The difference between spec-driven development and traditional approaches ✅ How to structure system use cases for effective AI code generation ✅ Why self-contained systems architecture is ideal for AI-assisted development ✅ Practical insights from real enterprise modernization projects ✅ How AI is changing team structures and release management ✅ The technical stack that works best for AI-driven Java development Key Timestamps: [You'll add these during editing] Guest: Simon Martinelli Java Champion | Vaadin Champion 🔗 linkedin.com/in/simonmartinelli 🌐 martinelli.ch 📚 unifiedprocess.ai Resources Mentioned: AI Unified Process: unifiedprocess.ai Self-Contained Systems: scs-architecture.org Vaadin Framework: vaadin.com jOOQ: jooq.org Karibu Testing: github.com/mvysny/karibu-testing Playwright: playwright.dev Subscribe for more conversations with developers pushing the boundaries of AI-assisted software development