Why a dinosaur is writing about AI
the dainosaur — chapter 1
I started working in IT on July 1st 1999. So I actually worked on-site at a bank on New Year’s Eve that year as part of a Millennium Bug rapid-response team. Luckily, we did not have anything to do that night. Now why would I start a blog-post series like this? Well, it is not to brag. It is context for everything that follows. It means I have lived through the dot-com boom and the crash, through ERP rollouts that cost fortunes and delivered mediocrity, through SOA, microservices, through the cloud, through agile transformations that somehow turned good engineers into people who spent half their week in meetings about meetings. Every single one of those waves arrived with a pile of whitepapers, a conference keynote with a hockey stick slide, and an army of consultants ready to sell you the future. My scar tissue has scar tissue. And with that comes something useful: a very sensitive bullshit detector.
So when AI started going mainstream, my gut reaction was not wonder. It was wait and see. Specifically: let us see what this thing can actually do before I get excited. Because I had a reference point. Many years of hard-won experience, patterns burned in from real projects, the kind of instincts you only build by making the call, watching it play out, and living with the result. I genuinely thought (and I am being honest here) that no language model was going to come close to that. A senior architect with a career’s worth of instinct under his belt versus a very confident text generator? I knew where I was putting my money. But I was wrong. More wrong than I expected, and in ways I did not see coming. That shift from skeptic to someone who is now all in, is what I want to write about. So do not expect tutorials, “top 5 prompts for architects”, benchmarks, tool comparisons or screenshots of things that impressed me last Tuesday. There is plenty of that content out there and some of it is genuinely good.
I want to address the actual experience of bringing AI into serious, senior-level work: the friction, not the features. The moments where it helps in ways you did not expect, and the moments where it confidently leads you somewhere stupid. The question of what you trust and what you verify, and how that changes when you have spent your career building your professional identity around knowing things. What happens to your craft when a tool starts doing parts of it? When does experience make you better at using AI, and when does it just make you slower to adapt? These are not abstract questions. They come up in real engagements, in real decisions, in the quiet moment at the end of a day when you wonder whether you actually added something today or just orchestrated a very expensive autocomplete.
I am writing this for people who are not new to the industry. If you are just starting out, welcome! But I will be expecting you to have a frame of reference that assumes years of experience, of building instinct, of watching things fail in ways nobody predicted. Architects, senior engineers, consultants who have strong opinions and enough self-awareness to question them. People who can see through a pitch but might still be curious about what is behind it. People who have something real at stake in how this plays out. Not just their career, but their craft. If you recognize yourself in that description, stick around.
Although not the main target audience of this series, I actually do often work with young people coming into the industry, and I genuinely enjoy it. Because they see things differently. They question things I stopped questioning years ago, and they are right to do so. One of my mottos is “you can learn something new from anyone at any time”, and I mean it. The flip side is that I have also made most of the mistakes they are still going to make. I know how some of these stories end. That combination of openness and hard-won pattern recognition is what makes a senior person actually useful rather than just expensive. When I look at myself in that mix, shaped by years of decisions and their consequences: I recognize the dinosaur. Not as an insult. But as an honest description. I am older than most of the stack we are running on, but still fully invested in AI. Hence, the title: the dainosaur. Dinosaur, AI, one word. The pun is obvious.
My goal is to be brutally honest about my journey into AI: the hesitation, the friction, the moments where I was wrong about something I was confident in. If you are stuck, or actively resisting AI, or somewhere in that uncomfortable middle ground where you can see it might be real but cannot quite commit: I have been there. I recognize that struggle. And I am not going to tell you it was easy or that the doubts were silly. They were not. But I have made the leap, and I am genuinely more productive than I was before. Not by erasing what I built over decades and replacing it with AI, but by combining the two. The experience and instincts stay. AI augments the work, it does not overwrite the person doing it. That distinction matters to me, and I suspect it matters to you too.
Next: chapter 2 – Letting go of the reins
“The person trying to tame the horse is fighting the animal. The person working with a skilled but opinionated colleague is having a conversation. Same situation. Very different results.”