I've been in technology for more than forty years. I've watched the industry move through mainframes, the rise of the web, mobile, cloud. None of it prepared me for the pace we're operating at right now.

This week alone, two things happened that I want to share — not because they're dramatic, but because they're the kind of ordinary working moments that explain where AI actually fits.

The Design Layer Just Collapsed

We've been working with Claude Design, a recent release from Anthropic, and the difference it's making in our workflow is hard to overstate. We pulled our repo directly into the tool. We brought in our Figma designs. And then, almost without realizing it, we removed an entire layer of complexity from our process.

We don't need to be Figma experts anymore. We don't need the handoff. The integration runs from design straight into our build, and we're shipping from concept to working interface in a fraction of the time.

For a team building an enterprise learning platform, that's not a small thing. That's the difference between "we'll get to that next quarter" and "let's just do it tonight."

When Microsoft Pushed a Platform Policy Change Mid-Workday

The second story is one I'm still grateful happened the way it did. Microsoft pushed a policy change to our Office 365 tenant in real time, one that hit how our team could log in. We happened to be logged in and working when it broke. If we hadn't been, things could have gotten ugly fast.

Normally we'd have to stop everything, start researching, and figure out what was actually going on. Instead, we reached for Claude. We worked through it together — Claude helped us trace what had changed, what was happening, and what to do about it. It was fast. It was accurate. It was spot on.

But here's the part I want to be clear about: it wasn't AI alone that solved it. It was AI plus decades of problem-solving and knowing what to look for. The two together got us to the answer in minutes.

We also came out of it better prepared. We understand Microsoft's auto-pushed changes far better now, and we've set alerts so we see the next one coming. That's the model I keep seeing work, and it's exactly why I get protective about how we build.

Why I Won't Call This Vibe Coding

I want to say this directly, because the term gets thrown around a lot: I'm not a fan of "vibe coding," and I don't want anyone confusing it with what we're doing.

For a simple website or a quick application? Vibe coding is fine. Have at it. But if you're building an enterprise platform that real people depend on — that your business runs on for revenue — you cannot get there by vibing. You need engineering. You need people who understand what the code is doing, who can troubleshoot when it breaks, who can maintain it for the long haul.

The code we write with Claude Code is more complex than what I'd write line by line on my own. I'll admit that. But do I understand what it's doing? Yes. Do my engineers understand it? Yes. That's the line. AI lets us move faster. It does not let us skip the thinking.

If you build something you can't support, you haven't built anything. You've built a liability.

What This Means for the People We Serve

This is why Nebula Academy exists, and why I keep coming back to the same point with the learners we work with: the win in this moment isn't about replacing your skills with AI. It's about layering AI on top of what you already know.

There are two gaps I see right now, and both are costing people real opportunities.

The first is the experienced professional — the person with deep domain knowledge in healthcare, finance, manufacturing, education, you name it — who's either afraid of AI or doesn't know where to start. That person is sitting on the most valuable raw material in the AI economy: real-world expertise. And they're letting it go untouched.

The second is the young adult walking into the workforce who feels threatened by AI. Worried it'll make them obsolete. Worried it'll make them feel less intelligent. That's also a missed opportunity. The young people who lean in, get uncomfortable, and learn to apply AI inside a business context are going to be the ones who define what these roles look like next.

Both groups are looking at the same door from different sides. And the answer is the same for both: combine what you already know — or what you can quickly learn — with the willingness to put in the work to understand how this technology actually functions.

The Real Skill Is Still the Skill

What I've been experiencing this week, and frankly every week now, is that AI is the best coworker I've ever had. It doesn't replace what I bring to the table. It amplifies it.

But the judgment, the architecture, the understanding of what should be built and why — that's still mine. That's still ours. And that's the skill set we're building at Nebula, because it's the one that will hold up no matter how fast the tools keep changing.

This pace isn't slowing down. The question for all of us is whether we're going to recalibrate forward and meet it, or stand still and watch it pass.

Build With AI the Right Way

This is exactly what we teach at Nebula Academy. Our CAPS (Certified AI Project Strategist) program shows individuals and teams how to build enterprise-ready applications with AI — on Microsoft Azure or any cloud service — and how to wrap that work in a full software development lifecycle. The result isn't a demo you can't support. It's something real you can ship, scale, and stand behind.

If this is the gap you're feeling — whether you're the experienced professional sitting on hard-won expertise, or the team that wants to move faster without building a liability — this is where you close it.

Explore CAPS at nebulaacademy.com

Laurie Carey is the CEO and Chief AI Officer of Nebula Academy, founder of We Connect The Dots, and author of Resilience Is a Muscle. She speaks and writes at the intersection of AI strategy and human resilience.