Welcome to the Blog: Continuing the AI Development Conversation

I’ve always been a generalist at heart, and as someone who collects physical books, I appreciate the permanence of a bound volume. But writing and releasing AI-augmented Software Development earlier this year presented an interesting paradox: I was taking one of the most volatile, rapidly accelerating fields in modern history and locking it into a static medium.

The core premise of the book—that we must view AI as a collaborative partner, co-worker, and co-thinker rather than just a tool—is foundational. The fundamental shift I wrote about isn’t going anywhere. The principles of a value-driven workflow, where we keep what truly matters like quality and maintainability while tossing out processes that only exist because of human limitations, will hold up for years to come.

I deliberately kept the book tool-agnostic. I wanted to ensure that the approaches and thinking patterns remain relevant regardless of which specific AI models or tools evolve. Concepts like the inverted development curve—acknowledging that AI development starts slower to build context but accelerates dramatically as projects progress—don’t depend on whether you are using Claude, Gemini, or OpenAI.

But theory needs practice, and the practice is changing at a breakneck pace.

When I wrote about how AI transforms our approach from typing to directing, the tools available required a very hands-on approach. We were just moving past the phase of AI as an advanced auto-complete. Now, just months later, the ecosystem is shifting violently again. The daily reality, the pricing models, and the friction of the developer experience are changing on a weekly basis.

A book provides the architecture; a blog provides the scaffolding.

I need a dynamic space to track how these concepts survive contact with the real world. While the book focuses on approaches that will remain relevant, this blog will be the exact opposite. It will be a place to get into the weeds of the ephemeral: the specific workflows, the current limitations of team collaboration, and the raw economics of running these tools day-to-day.

Expect field notes, architecture teardowns, and ongoing experiments as the math of software development continues to change. Let’s see where this goes.