Beyond Stability: Embracing Change in System Design

Beyond Stability: Embracing Change in System Design

I’ve helped build many systems over the years—most of them are still running, still in use, still evolving. I don’t have a formal certification like TOGAF. I’ve rarely drawn an Entity Relationship diagram that followed textbook conventions.

This isn’t a rejection of structure—it’s a commitment to relevance. It’s like the stories of founders who dropped out of grad school: not because education doesn’t matter, but because their obsession with the problem overtook their need for credentials. For me, real systems thinking starts with empathy for the problem, not the diagramming of it. The methodology flows from the mission—not the other way around.

The business context is everything: the organisation’s culture, history, market dynamics, customer expectations, and internal politics. You can’t design good systems without feeling that context. You don’t impose structure—you grow it.

Now add AI into the mix, and the role of the programmer is shifting fast. AI can now handle many of the mechanical aspects of software development: generating apps using best practices, writing test cases, scaffolding architecture. What it can’t yet do is care. It can’t interpret a messy human domain and say, “Here’s what actually matters.” That’s our job.

Programming is no longer just about code. It’s about managing information flows, designing for traceability of change, and crafting interfaces—both technical and human—that allow systems to evolve safely. It’s about understanding where stability is needed and where flexibility must be preserved. And it’s about recognising that reuse, modularity, and abstraction aren’t virtues in isolation—they’re context-dependent survival traits.

“The best architectures don’t come from abstract ideals—they come from being deeply embedded in the business context.”

With AI accelerating the routine parts of development, our role shifts. We must lead teams that don’t just write code—but understand intent, navigate complexity, and design for evolution.

Originally published on LinkedIn.

Share: X (Twitter) LinkedIn