You know the exact feeling. You type out a beautifully structured prompt, hit enter, and wait. And wait. In the old days, you’d be in the zone, fingers flying across the keyboard, holding the entire architecture of the application in your head. Now? You’re staring at a spinning cursor, your flow state shattered into a thousand tiny context switches.
We were promised an amplifier for human skill. Instead, we got a cognitive crutch. The tool designed to amplify your skill is quietly outsourcing your cognition. You can literally feel your neurons clogging up as you wait for the AI to finish thinking, forcing you to stare blankly out the window for ten minutes just to reset.
It’s not just about focus, though. It’s about ownership. There used to be a time when a client reported a bug, and you immediately knew exactly which file to open and which line to fix. You understood the DNA of your project. Today? You write extensive AGENTS.md files, craft meticulous design docs, and then clasp your hands to recite a prayer to the LLM deity, hoping it understands your intent.
When the AI goes down—and it will, every other day—you are paralyzed. You don’t understand 80% of the codebase because you didn’t write it. You don’t own a codebase you can’t debug without a WiFi connection. You have traded your resilience for convenience, and the price is your independence.
And the worst part? The AI is mostly right, but never quite there. It does 80% of the work flawlessly, but the remaining 20% is a labyrinth of wrong assumptions. You spend hours holding its hand, fixing bugs that it reintroduces just as quickly as it solves them. It’s Penelope’s shroud—you make progress during the day, only for the AI to unravel it at night. No real ground is ever gained.
People will tell you that hating coding agents is just a “skill issue.” They’ll say you just need to learn how to prompt better. But the real skill issue isn’t about prompt engineering. It’s about recognizing when a tool is eroding the very craft it claims to elevate. We are trading the deep satisfaction of mastery for the shallow convenience of a prompt. If we aren’t careful, we won’t just be worse engineers—we’ll be mere operators of machines we no longer understand.
FAQ
Q: Isn't this just resistance to progress, like when developers hated compilers?
A: No. A compiler translates your exact logic into machine code; it doesn't invent the logic for you. Coding agents are asking you to outsource the actual problem-solving, which is the core of engineering.
Q: So I should just stop using AI completely?
A: No, use it for boilerplate, refactoring, or quick syntax lookups. But if you're using it to architect your project and write core logic, you're actively degrading your own expertise.
Q: What's the contrarian take?
A: The AI isn't making you a worse engineer; it's exposing the fact that many 'engineers' were just glorified stack-overflow copy-pasters all along. The AI simply automated their entire skill set.