How I Build Software

AI-assisted development is here to stay. The question is no longer whether teams should use AI, but how to use it effectively.

Many organizations are discovering that simply adopting larger models or paying for more expensive AI tools doesn't automatically lead to better software. Productivity comes from engineering discipline: providing the right context, establishing clear constraints, and creating fast feedback loops.

Context Matters

Even affordable open-source models can produce excellent results when they understand the problem they're solving. Business requirements, architectural decisions, coding standards, and domain knowledge are often more important than the model itself. I focus on building systems that provide AI with the context it needs to make informed decisions.

Feedback Loops Matter

AI performs best when it can validate its work. Strong typing, well-defined interfaces, unit tests, integration tests, and automated checks help both humans and AI converge on correct solutions faster. The goal is not to generate more code, it's to generate better code.

Engineering Still Matters

AI can accelerate implementation, but it cannot replace engineering judgment. Generated changes should be focused, understandable, and aligned with the long-term health of the codebase. Knowing what to keep, what to reject, and what to simplify remains a fundamentally human skill.

I take pride in being able to build software with or without AI. The tools change; good engineering principles do not.