Why AI Coding Assistants Misunderstand Your Architecture

AI coding assistants misunderstand your architecture — not because they’re buggy, but because architectural intent was never visible in the code to begin with.
In this video I read through and comment on Ian Bull’s article “Sinks, Not Pipes: Software Architecture in the Age of AI”, adding my own perspective from my work on usable software design. The argument that structuring code for comprehension has real economic value — one I’ve made for years — is now impossible to ignore.

We cover:

• Why AI agents permanently live in the gap between architectural intent and code structure
• The sinks vs. pipes distinction — why cascading side effects are fatal to AI-assisted development
• Deep modules, low coupling, high cohesion — why the old principles matter more now, not less
• Practical steps: ADRs, architectural tests, context artifacts

The problems aren’t new. The cost of ignoring them is.

Original article by Ian Bull: https://ianbull.com/posts/software-architecture