Member-only story
Java Developers Are Learning the Wrong Skills for the AI Era. Here’s What I’d Learn Instead.
AI can write your Spring Boot controllers, generate your repositories, and pass your unit tests. Unfortunately, production doesn’t give a damn how quickly you wrote the code.
A Java developer recently asked me whether he should abandon Spring Boot and learn Python because of AI. He had spent three years building backend applications, understood REST APIs, could write a respectable amount of Hibernate code, and had just watched an AI coding assistant generate something that would have taken him an entire afternoon.
He wasn’t worried about Python. He was worried that three years of experience had suddenly become a fifteen-dollar monthly subscription.
I understand the concern, although I think he’s looking at the wrong part of the problem. AI is becoming extremely good at producing the kind of Java code that fills most beginner tutorials, corporate onboarding exercises, and approximately half the pull requests nobody particularly enjoys reviewing.
Give a coding assistant a specification for a Spring Boot CRUD application, and it can generate controllers, services, repositories, DTOs, validation…
