Real Talk: You Don't Miss Engineering. You Miss Typing

An anonymous engineer argues that AI-driven code generation is the next rung on the same ladder as compilers, high-level languages, and package managers, and that engineers who fear losing the act of typing are confusing nostalgia with engineering. The engineer contends that the real program lives in engineers' heads rather than in source code, and that as code generation becomes cheap, verification and design judgment become the scarce skills. The piece urges engineers to embrace working alongside AI while warning against "vibe coding.

As a kid, when I watched sci-fi movies about computers with human-like thinking, I was completely sure they'd never exist. That was in the early 2000s. Now we live alongside them. Engineering isn't ending. It's starting. Lately I keep seeing engineers declare that "real engineering is gone," that "they're done with this s...," that "it's all AI now," and that they miss typing every line themselves. I understand the feeling, but I think it confuses nostalgia with how engineering and science work in general. I feel bad for engineers who can't see that this is one of the best times, if not the best time, to be an engineer. Look around: things change almost overnight. For most of human history, life changed slowly, with generations living in roughly the same world as their parents, grandparents, or great-grandparents. We're living through one of the most dynamic eras ever. Sure, I still have a soft spot for '80s vibes, even though I wasn't even born yet. But there's no need to invent a problem and suffer over it when there are so many things to catch up with. Just ask yourself how many things you don't know that you want to know. Tech and science don't give a damn about the good ol' days. It's always been like this. When Grace Hopper built early compilers, skeptics doubted "automatic programming" was serious work. When FORTRAN arrived, programmers doubted compiled code could match hand-written assembly. High-level languages, frameworks, and package managers all triggered the same fear, and each time the industry grew. AI is the next rung on the same ladder. Never. You spend exponentially more time thinking about the problem than typing the code that expresses your thoughts. The real program is the understanding in the engineers' heads, not the source code. AI can generate thousands of lines, but someone still has to hold the theory of why the system works the way it does. Speeding up one step only helps as much as that step mattered. Make code generation instant, and the constraints shift to requirements, design, review, and debugging. Generating code is getting cheap; verifying it is not. That judgment is engineering. Every time software got cheaper to build, we built far more of it. Lower costs unlock problems that were never worth solving before. What matters is who does the engineering: you, the AI, or both of you together. My bet is on the engineers who choose both and keep the thinking for themselves. So please, just stop dwelling on things that are inevitable and focus on the real problems, and the real opportunities will show up. Just don't vibe code, please That's not engineering and nobody can change my mind about that. I hope this little note helps a bit. Enjoyed this brief note? Let's stay connected I share more software engineering insights, projects, and experiments across these platforms: