The gap AI agents fall into
When you ask an AI agent to "add rate limiting to the API" and it will produce something. It might even compile, pass a quick test, and look complete in the diff.
But what does "rate limiting" mean here? Per user, per API key, per IP? What happens when the limit is hit — a 429, a queue, a silent drop? Is this a business rule with a specific number att...
[Read More]
Why Do I Need to Use the Parallel Change Pattern?
The hidden risk in fast code generation
AI coding tools are very good at producing a complete-looking change from a short request.
That is useful when the change is local. It becomes risky when the change crosses compatibility boundaries: database schemas, public APIs, event contracts, configuration keys, command outputs, generated artifacts, or workflows consumed by other services.
A human enginee...
[Read More]
From Code Generation to Software Engineering: Why AI Agents Need Design Skills
The shift from generation to engineering
The first wave of AI-assisted development was mostly about speed. Ask the model for a class, a test, a migration, a controller, a README section, or a refactoring, then review what came back.
That capability is useful, but it is not the same as software engineering.
Software engineering is not only the ability to produce code. It is the ability to change a s...
[Read More]