Treating Architecture Models as Living Context for AI-Driven Development

Software engineers have long treated architecture diagrams as static documentation that quickly falls out of date as codebases evolve. With AI coding assistants and autonomous agents now generating implementations, maintaining accurate architectural context has become critical to ensuring software aligns with design intent and security requirements. Specification-driven development tools like AWS Kiro have helped by embedding structured requirements into repositories, but they still fall short of capturing how individual features fit together architecturally. One engineer has been exploring a different approach through a project called iSecureByDesign, which treats architecture models as a dynamic source of truth capable of generating specifications, security controls, and implementation guidance. The experiment proposes a shift from architecture-as-documentation to architecture-as-executable-knowledge that actively participates throughout the software development lifecycle.