Articles on spec-driven development, requirements engineering, and AI-ready architectures
User stories are popular in agile teams. They are short, readable, and focused on user value. For many teams, they help with planning and coordination. However, when the goal is Spec-driven Development, user stories show serious limitations. They push teams to create plans and task lists before the real requirements are clear. This makes them a
Read ArticleSpecifications in spec-driven development must be readable by all stakeholders, not just technical teams. When business participants can understand the spec, errors surface earlier, assumptions decrease, and trust grows across the project.
Read ArticleUser stories alone often lack context. Use-Case 2.0, created by Ivar Jacobson and others, groups related stories under unified goals and delivers the structural clarity that AI-assisted development needs.
Read ArticleTools like Amazon Kiro, GitHub Spec Kit, and BMad Method promise structure for AI-assisted coding but struggle with brownfield enterprise systems. The AI Unified Process takes a fundamentally different approach better suited to existing codebases.
Read ArticleThe term "use case" is widely used but not always consistently. This article distinguishes business use cases focused on organizational behavior from system use cases defining system behavior, and shows how both contribute to AI-assisted requirements engineering.
Read ArticleCurrent AI tools focus too narrowly on code generation for developers. The real enterprise software challenge lies upstream in clarifying requirements, defining specifications, and achieving stakeholder alignment.
Read ArticleCompares two approaches to AI-assisted software development. While BMAD focuses on orchestrating multiple AI agents, clear specifications form a more effective foundation than complex agent workflows.
Read ArticleAI has made coding faster and cheaper, but the real challenge has shifted to understanding what systems should do. Clear requirements are now the critical bottleneck.
Read ArticleExplores building integrity into systems through conceptual and perceived integrity, requiring excellent information flow, the Chief Engineer model, and robust technical practices.
Read ArticleAI is shifting software development focus from code to specifications. Introduces ReDevTest, where clear requirements and acceptance tests take precedence over implementation.
Read ArticleTrue spec-driven development prioritizes stable specifications describing intent and system behavior over fragile task lists, enabling long-term sustainability over short-term productivity.
Read ArticleExamines whether use cases and user stories truly contain identical information by comparing them through real, non-trivial examples.
Read ArticleComparing deterministic and iterative approaches to spec-driven development, and why iterative wins for real-world projects.
Read ArticleSpec-driven development doesn't mean waterfall. Learn how it fits naturally into agile, iterative workflows.
Read ArticleSystem use cases provide the structured, unambiguous specifications that AI needs to generate reliable code.
Read ArticleAI makes developers faster at coding, but the real bottleneck is elsewhere. AI Unified Process makes the full impact visible.
Read ArticleHow the IREB AI4RE micro-credential aligns with AI Unified Process principles for AI-assisted requirements engineering.
Read ArticleWhy jumping straight into code is the wrong approach and how system use cases create a better foundation for development.
Read ArticleDesigning application architectures that are optimized for AI-driven code generation and maintenance.
Read ArticleSelf-contained systems align perfectly with AI-driven development by keeping context manageable and boundaries clear.
Read ArticleThe foundational article introducing spec-driven development with AI and its roots in proven software engineering methodologies.
Read ArticleArticles on Vaadin, jOOQ, and why they are ideal for AI-driven development
Traceability connects code to business requirements. This article explains why tracing use cases to tests matters, demonstrates the annotation-based approach used in the AI Unified Process, and introduces the AIUP Navigator IntelliJ plugin for navigating between specifications and test code.
Read ArticleShowcases a feedback application built to demonstrate the AI Unified Process methodology, emphasizing that effective AI-assisted development prioritizes specifications and clear system behavior before code generation.
Read ArticleHow Vaadin's server-side Java UI and jOOQ's type-safe SQL combine to create an ideal stack for AI code generation.
Read ArticleVaadin's pure Java approach eliminates frontend complexity, making it the ideal framework for AI-assisted development.
Read ArticleAI Unified Process combines the best of Rational Unified Process with modern AI tooling