-
Learn a repeatable framework for reliable AI-assisted development
-
Build a practical full-stack application using spec-first workflows
-
Get expert guidance from AWS and software architecture specialists
AI coding assistants can dramatically improve developer productivity, but without structure, they can also produce inconsistent, hard-to-maintain code. Spec-Driven Development gives developers, engineering leads, and product teams a disciplined way to guide AI tools with clear, human-readable specifications before writing code.
In this hands-on workshop, Ricardo Sueiras, Principal Developer Advocate at AWS, and Simon Martinelli, Programming Architect at Martinelli LLC, show you how to move from chaotic prompting to repeatable, production-ready AI-assisted development. You’ll learn how to turn vague ideas into EARS-formatted requirements, break specifications into executable tasks, and guide AI coding assistants through structured development phases.
The workshop also explores how Amazon uses spec-first engineering practices to build reliable systems, and where tools such as Amazon Kiro and Amazon Q fit into modern development workflows. Through a practical full-stack build, you’ll gain techniques for improving AI accuracy, reducing rework, and keeping specifications, design, and code aligned.
By the end of the workshop, you’ll be able to apply Spec-Driven Development to build more maintainable, testable, and scalable software with AI coding assistants.
This workshop is for developers, Python developers, AI product managers, team leads, and engineering leads who want to use AI coding assistants more effectively. Intermediate programming experience is recommended. Examples will use Python, but participants can follow along using their preferred programming language.
-
Write clear EARS-formatted requirements
-
Turn vague ideas into precise specifications
-
Break specs into AI-executable tasks
-
Guide AI through structured development phases
-
Improve AI accuracy with context engineering
-
Keep specs, design, and code aligned
-
Use tools such as Kiro and Amazon Q effectively
-
Build a hands-on full-stack demo application