# Should you learn .NET if your job doesn’t use it?

With AI writing more and more code, it’s fair to ask whether learning another language is still worth the time. I think it is. I’m biased, having spent years building fintech, insurance and lending platforms on .NET, but I’d make the case even if I weren’t.

AI can help you write C# even if you’ve barely touched it. But there’s a big gap between generating code and understanding what it does, whether it’s any good, and how it will behave in production.

Learning .NET pushes you to understand things that matter in any stack. How the async/await model works and what really happens when you block on it. How generics, nullable reference types and a strong type system catch mistakes before they ship. How the garbage collector works and why allocation rate matters. How dependency injection and middleware shape an application. Those fundamentals travel well beyond .NET.

There are other good reasons too. If you enjoy understanding how software works in detail, it’s a different way of approaching problems. It also opens career options, since .NET runs a great deal of banking, insurance, fintech, enterprise software and Azure workloads, and it powers game development through Unity. And there’s a real trend to get ahead of: many companies are still moving older .NET Framework and early .NET Core systems onto modern .NET, and engineers who know the current platform are well placed when those teams hire.

If you already work in Java, TypeScript or Python, you have skills to build on. C# will feel familiar to Java developers. TypeScript developers will recognise the async/await model, and the two languages share a lead designer in Anders Hejlsberg. Python developers will find LINQ a natural cousin of comprehensions. You don’t need to start from scratch, and a small [ASP.NET](http://ASP.NET) Core API as a side project is enough to begin before using it professionally.

AI is changing how we write software, but understanding the code, making good technical decisions and knowing when something isn’t right still matter. Strong fundamentals become more valuable when AI does more of the implementation.

For engineers in my network, has AI changed how you decide which languages to learn?