It’s a strange time to be a developer.
Not bad — just different. You scroll through Twitter (sorry, X) and suddenly everyone’s a “prompt engineer.” People are shipping full-stack apps with three sentences and a screenshot. Job listings ask for LangChain before they even mention JavaScript. And you start wondering… do I even need to write code anymore?

Prompting: The New Way to Build

A few years ago, “prompting” meant CLI flags or asking a user for input. Today, it means crafting the perfect sentence to get GPT to build a microservice — complete with error handling, tests, and documentation. I’ll admit, it’s incredible. I’ve used prompts to scaffold projects, generate test suites, refactor messy legacy code, and even write bash scripts I didn’t feel like Googling. It works. It’s fast. It’s fun.

But it’s also… odd.
When you start writing prompts instead of functions, you stop exercising the problem-solving skills that drew you to programming in the first place.

From Developers to API Wranglers

Something has shifted. Not just the layoffs — the whole feel of being a dev. It’s less “crafting software” and more “assembling pieces from models and APIs.” You write some glue code, prompt an LLM for a function, paste in a Stack Overflow snippet and have ChatGPT “clean it up,” then ship. Done.

Is that bad? Not necessarily. Tools evolve. Abstractions pile up. Nobody codes in assembly anymore (unless you’re very, very cool… or cursed).

But the vibe is different. I don’t get the same satisfaction from writing a clever prompt that generates a Stripe webhook handler as I did building it by hand. I miss zoning out in VS Code. I miss poring over documentation. I miss reasoning through edge cases in my head — instead of asking AI to “handle edge cases” and hoping it gets it right.

Why Coding Still Matters

Prompting is powerful, but it’s not magic. When something breaks, needs to scale, or runs into a gnarly race condition, you need to understand what’s actually happening.

Knowing how to code is still the real superpower. The prompt is just a shortcut.
It’s like autopilot in a car — great for the highway, but if you can’t parallel park when the system glitches, you’re in trouble.

I’ve seen this play out on teams: new devs who only know how to prompt hit a wall with debugging, testing, or building anything non-trivial. Senior engineers are still the ones solving the last 10% of problems AI can’t handle.

But Don’t Be a Dinosaur

Refusing to adapt is a fast track to irrelevance.
I’ve seen developers mock “prompt engineers” as if they’re not real coders. The same way others once mocked front-end devs. Or people who used Rails generators. Or those who didn’t write in C. See the pattern?

Prompting is programming now — just another layer in the stack. Ignore it, and you’re ignoring a powerful abstraction that can make you faster, more creative, and more productive. The real skill is knowing when to use AI… and when to be the AI.

What We Might Lose

Maybe creativity.
When AI writes the code, our role shifts from creating to curating. From making to editing. It’s like going from painting to photo retouching — still art, still skill, but… different.

When everything looks like GPT output, apps start to feel the same. The little quirks, the human fingerprints, fade away. And sure, maybe that’s fine for CRUD apps. But the best software — the kind that feels unique — usually comes from human weirdness, not machine predictability.

Final Thoughts

Prompting is here to stay, and programming will never go back to what it was. The line between “developer” and “AI wrangler” is already blurry. If all you do is prompt, you’ll lose the soul of building.

So keep learning to code. Keep building. Use AI, but don’t let it replace your curiosity. Ask “why,” not just “how.” Remember — the real creativity comes from the mind behind the prompt, not just the box it’s typed into.