Blind trust in AI can be as perilous as Michael Scott trusting the satnav (opens in new tab) so blindly that he ended up driving right into a lake. Back in the early days of MySpace, I spent countless hours tinkering with code, manually tweaking styles to get the look I wanted. It was messy, hands-on work. But it taught me the basics: how to debug, how to fix problems when things went awry. Today, many developers leap into projects with AI-generated code without understanding the fundamentals. Without those basics, debugging and problem-solving become near impossible, especially when AI blunders.

Yet, AI isn’t without its benefits. It simplifies complexity, reducing the cognitive load that comes with understanding intricate systems. But there’s a catch. If you let AI do all the heavy lifting, you risk becoming complacent. You might stop learning the underlying principles, losing the ability to understand why something works or fails. This becomes a significant limitation in the development process, and it’s one I’m also guilty of sometimes. Imagine a site that breaks unexpectedly. AI might fix it, but without understanding why it broke, you can’t prevent the issue from reoccurring, and you might have spent thousands of tokens trying to fix it.

To quote Andrej Karpathy from a post (opens in new tab) on XCancel

They will implement an inefficient, bloated, brittle construction over 1000 lines of code and it’s up to you to be like "umm couldn’t you just do this instead?" and they will be like "of course!" and immediately cut it down to 100 lines.

Curiosity is your ally here. Stay inquisitive about how things work, even as AI offers shortcuts. Don’t let its simplicity lull you into a state of laziness. Understanding the "why" underpins everything from UX design to backend development. You don’t need to be a database guru, but you must grasp the implications when your site or app fails or performs flawlessly. This foundational knowledge is what sets skilled design engineers apart.

AI’s limitations are stark. It’s fundamentally non-creative, selecting the most statistically likely next token based on its training data. This means it defaults to the common denominator, those most common UI patterns and predictable outcomes. Designers, however, thrive on these nuances. We appreciate the fine details, the intentional choices that aren’t dictated by statistical likelihood. That’s where our expertise lies.

Design engineers hold a unique position today. We marry the designer’s eye for detail with the engineer’s ability to execute. This combination is our current advantage. AI might level the playing field, but those who understand the "why" still set the "ceiling."

Remember: AI complements rather than replaces human creativity and expertise.