← Notes
AI-Native Engineering · Agentic Software Engineering · Software Architecture
Using AI coding agents without losing architectural control
A practical note on treating AI-native engineering as an extension of software-engineering discipline, not a replacement for it.
Software development is changing, but production responsibility has not disappeared.
My AI-native workflow uses coding agents and large language models for research, codebase exploration, implementation planning, refactoring, testing and delivery. The important distinction is control: architecture, verification and production judgment remain engineering responsibilities.
The goal is not to generate more code. The goal is to shorten the distance between understanding a problem and shipping the right solution.