Let me ask: in what way have you been proven wrong?

Have you simply just seen it build a pretty CRUD application that you feel like would take you many hours to do on your own? If that's the case, then you really shouldn't be too worried about that. Being able to build a CRUD application is not what you're getting paid for.

If you've seen something that involves genuinely emulating the essence of a software engineer (i.e., knowing what users want and need), then I'd ask you to show me that example.

Anything that involves 1) Physical work, 2) Legal requirements and 3) Union should keep you safe for a while.

And if the world devolves into a more chaotic one, I bet they will accelerate the advancement of AI.

I don't think there truly is a way to bulletproof yourself

Anything that AI can't do today could be convincingly argued that it will be able to do in 3 years. There's no-one that can tell you what the world will look like then

If you're seriously worried and can't take that risk then maybe look into switching careers

Learn to sell yourself. Learn to adapt quickly. Learn to learn

The days of having one skillset you learn deeply and monetize for 40 years are over

(Yes I know the trend I’m about to talk about was different in BigTech and adjacent. I did an almost 4 year stint at a FAANG 2 years ago and have no need or desire to go back at 51. That’s not where most developers work)

Back around 2013-2014 well before AI, I saw the trend that it wasn’t going to take much to be a “good enough” enterprise “full stack” developer meaning the market was going to be commoditized and if i was just a “I codez good and pull tickets off the board” developer, it was going to be hard to stand out from the crowd.

I was prescient, when companies reach out to me now for standard enterprise dev jobs in Atlanta (no longer live there but a large part of my network is still there) I see the same salaries I saw in 2016.

I worked over the next six years to move “up the stack”. I learned soft skills, learned AWS and how to treat it more than just a glorified Colo, learned how to lead projects, talk to the “business”, focused on business value, started getting closer to sales and how they operate etc.

Out of everything I said, the take away should not be to “learn cloud”. Everyone and their dog knows AWS (except for one niche that has opened a few doors to me).

The take away is to incease the “scope and impact” of your work beyond just being a code monkey. Also learn how to work at an increased level of ambiguity.

https://www.levels.fyi/blog/swe-level-framework.html

AI is just another part of the commoditization and devaluation of the generic developer. There is no “moat” around being a generic full stack/mobile/web developer.

As of 2020, I work in cloud consulting specializing in app dev. I have been working full time for consulting departments (the internal one at AWS) and after being Amazoned in late 2023, it took me all of three weeks to have three offers. After I made a bad choice between the three in 2023, it again took me two weeks to have an offer.

I haven’t had to do a coding interview in over a decade because I know how to communicate my value and experience and I don’t try to compete on “I codez real gud”. I’m not a super special snowflake