Spanning Change: Software Engineering in the AI Era

The AI tools are getting faster. Are you getting better—or just getting busier?Every experienced engineer has had the moment: you paste a stack trace into a chat window, get a useful answer in seconds, and quietly wonder whether something fundamental just changed. It did.Generative AI isn't the next productivity tool. It's a general-purpose technology—in the same category as electricity and the internet—and it's compressing years of capability shifts into months. Engineers who figure out how to work with it well are pulling ahead. Engineers who are waiting to see how it shakes out are falling behind, slowly and then all at once.But there's a trap on the other side. If you treat the model as a source of truth—accept its outputs uncritically, let it do your thinking, ship code you can't fully explain—you're not becoming an AI-augmented engineer. You're becoming a passenger. When something breaks at 2 a.m., you'll be staring at a system you no longer own.Spanning Change is written for engineers who want to navigate the middle path: leaning into these tools without losing the judgment that makes you valuable.What you'll find inside:A clear-eyed framing of why this moment is different from previous AI hype cycles—and what makes it a genuine platform shiftA working mental model of how LLMs actually behave: next-token prediction, context windows, hallucinations, and agents—explained for practitioners, not researchersSeven attributes of engineers who are genuinely thriving with AI tools, plus repeatable techniques you can apply starting Monday morningHow to bring AI-augmented practices to a team without starting a process revoltThe emotional realities of a platform shift: FOMO, the temptation to coast, and what to do when someone less experienced is pulling ahead of youWho this book is for: If you've been shipping software long enough to have survived at least one platform shift, this book was written for you. The practices here don't assume any particular tool or level of AI adoption at your organization—they're grounded in how software engineering has always worked, augmented by what's now possible. If you're earlier in your career, everything still applies—you'll read it as preparation rather than recognition, and skip the years of unlearning that come with having done it the old way.What this book isn't: It's not a prompt-writing manual, an ethics treatise, or a prediction about which model will win. It's a framework for engineers who want to stay sharp, stay relevant, and still own what they ship—through this transition and the next one.Steve Recks has spent over thirty years shipping software in financial services, building and maintaining systems where correctness isn't optional and platform shifts don't wait for you to get comfortable. He's navigated enough technology transitions to know the difference between a cycle and a genuine shift, and enough engineering reality to be skeptical of both hype and hand-wringing. He writes from the middle of this transition, not from the other side of it.The faster the tools get, the more the judgment layer becomes the load-bearing part of the work. The engineers who understand that — and build accordingly — are the ones who still own what they ship five years from now. Read more