My web design degree taught me to build sites by hand, one piece at a time. What's actually different now isn't that I stopped needing to understand what I'm building. It's that I can describe what I want and get real, working code back in minutes instead of days, and still be the one who has to know whether it's actually right.
What building a site used to require
I picked up my associate's degree in web design in June 2013, and it taught me the real mechanics: structure, layout, how a site actually functions under the surface. That part of the education did its job well. It also meant every effect, every animation, every piece of interactivity had to be built by hand, one line at a time, which takes real hours even once you know exactly what you're doing.
What replaced that, on this exact site
On August 27th, I rebuilt marieharvey.com from a plain static page into a much heavier, more technical build: a scroll-driven hero with smooth scrolling and real animation throughout the site. That's a meaningfully more advanced build than the original plain HTML version, and it came together in a fraction of the time hand-coding those same effects alone would have taken me.
Proof this isn't just theoretical
The clearest example on this exact site is the link-in-bio hub page I use for Instagram and TikTok, rebuilt this month into a short scroll-driven film instead of a plain list of buttons, complete with a drag-to-rotate element and its own scroll-triggered animation. Something at that level of polish, built and shipped in days rather than weeks, simply wasn't realistic for me to hand-code alone on top of running two businesses and raising seven kids.
Does that mean I don't need to understand the code anymore?
No. If anything it's the opposite. Every real bug that's come up in a build like this, and there have been several, only got caught because I was actually looking at what came out, testing it for real, and pushing back when something looked or worked wrong. Building faster doesn't mean building blind. It means I get to spend my attention on catching real problems instead of typing out the boilerplate around them.
Why I didn't move this site to WordPress instead
I've stuck with a static site here, on purpose, and I'm planning to keep it that way indefinitely. WordPress has always felt slow to me, and I don't need its editing interface anyway, since the content gets written directly into the actual code instead of through a separate dashboard I'd have to learn and maintain.
The day I almost moved this site to WordPress anyway
On the exact same day I decided to rebuild this site into its current, heavier version, I actually considered migrating it to WordPress instead, then reversed that same-day call before it went anywhere. Being able to stand up a real comparison fast enough to make a decision like that in one sitting, instead of over weeks of research, is its own small example of how much faster real back-and-forth with AI makes decisions like this one, even the ones that end with staying put.
The one part AI never touches on this site
Every photo and video across this entire site, including the ones illustrating these posts, is a real one of me or my actual work. None of it is AI-generated, and that's a deliberate line I hold everywhere else in these businesses too. AI changed how fast I can build the container a site lives in. It never gets to decide what a visitor actually sees of me inside it.
What this makes possible that wasn't there before
This site now has its own standalone repository that a Git-based deploy tool on my cPanel hosting account pulls from automatically every time I push a change, landing straight in the real folder my domain actually serves from. That means I can genuinely ship a real update to a live site myself, without needing a developer on retainer for every small change.
The other half of this story, the marketing gap the degree never covered, is in what my web design degree actually taught me.