And yet...

I have my doubts, and some concerns about what the future holds with the way some things are going. A tinge of concern that a cyberpunk dystopia might just be in our future has started to creep in around the edges.

Because developing software costs so much money, and requires taking a lot of risk -- because software development lacks the predictability and repeatability of "real" engineering, see my other comment -- corporate executives are always looking for ways to reduce those costs and get something they can plausibly call "working software." Those execs have tried no-code/low-code several times. COBOL and SQL both promised end-user coding, with the usual result, because learning a language is the easy part of programming. Offshore outsourcing made a dent for a while but demand increased faster than supply of programmers, and now the offshore places have their own domestic demand for programmers. The current thing is LLMs, what we're calling AI these days. The VCs and companies who have a lead and a lot to gain are selling a big pile of hype to corporations. And some of it may work, I don't know. Based on history I expect LLMs to displace people in lower-end rote jobs, and that will likely include at least some of the easiest kind of programming. If VSCode can write 75% of the code for you and underline the errors an LLM can probably do even better.

I started in my career in the late '70s, and have stayed consistently employed since then. I never planned to make programming a career (I studied history at school and my parents hoped I'd go to law school), but it was challenging, came natural to me, and paid really well, so I've stuck with it (and now I can't do anything else to make a living). But I can't really put myself in your shoes, or relate to the OP's concerns except in an abstract way. I don't apply for jobs or go through hellish (and useless) interviews, although I read about those. I don't see LLMs taking over my job in what's left of my working years, but I suppose it could happen.

I'd like to tell you to sit out the hype cycle, the AI/LLM bubble will burst, the hype won't turn into a workplace revolution, the jobs will come back. I don't know that for sure but that's what I would bet on. But that doesn't do you or anyone else any good, looking for a job today, in a terrible job market full of recently laid off people who have at least some experience and (if they paid attention) some professional contacts.

Friends and professional contacts -- real people I know, not connections on LinkedIn -- have always worked best for me, so I advise meeting people and making friends, both in tech and in other fields. I have got some surprise job leads and freelance opportunities just chatting with people. I make it a habit to talk to strangers (to keep my natural shyness in check), you can try that. Word of mouth really works. If I had to find work today I would be talking to everyone I know, people I worked with 20 years ago, not filling out online job applications. I understand that's harder for someone without a lot of work experience.

As I wrote in my original comment, employers always need people who can solve business problems and add value. They don't always need more programmers. Figure out how to present yourself as a person who can solve problems, figure things out, work with the organization, add value, and you will find jobs, because people who can do that are always needed. It just isn't that rare anymore to have the ability to crank out web app code, and LLMs will make that meager skill even less valuable.

If you can speak and write effectively people will tell you, I guess that's the benchmark. If you can keep someone's attention, persuade someone to change their mind, communicate your skill and see someone use that knowledge, that's effective. You certainly can learn to speak and write better. Try joining Toastmasters to get over fear of speaking to groups. Watch videos of good presenters and speakers, or go see them in person. Read books about good writing, read novels that have held up for a long time (classics), and read essays and books by people who love reading and writing (Umberto Eco comes to mind). Go to writing classes and workshops.

Communicating well, and having the ability and knowledge to talk about more than one thing, will make you more appealing to other people, including potential employers and customers. I have freelanced for over ten years, and the number one thing I hear from new customers -- almost without exception -- is how the last people they hired stopped communicating with them. It doesn't take much effort to acknowledge and answer an email or call, but too many people fail even at that.

I like to say that maybe there's no such thing as the 10X programmer, or the 10X communicator. But you can easily observe that most people are at 0X or just barely above that. Be at least the 1X, worry about 10X later.

As for fundamentals, I don't know how to show that directly. A lot of it comes from practice and mentoring from more senior people, at least in my own experience. I'll repeat that if you can solve actual business problems and add value you will have a big advantage. You show, don't tell. Knowing the fundamentals will let you show and perform with confidence because you will see the same problems over and over, and if you already know how to approach the problem your employer or customer will understand that you know the fundamentals.

I know that many companies have internal designations tied to pay grades, like SWE II and so on. Other than the implied higher pay and more responsibility I don't think those titles mean much either, because they vary too much across companies.

If you mean can we credibly call programming an engineering discipline, personally I don't think so, but I have friends who argue the other side. I think of engineering as a trade that has established standards and practices so every new batch of juniors doesn't have to start from scratch and make the same mistakes. Those practices should allow engineers to reliably, predictably, and repeatedly make something from a plan or specification. I'm sure I could find exceptions, but most of what gets called software engineering has none of those attributes. We can't predict or estimate software projects accurately, not even close. We can't promise a reliable result, or even that the result will meet requirements. Even if the project succeeds we can't confidently repeat the performance, or show another team how to succeed.

I did some work in the automotive industry. The engineers who design cars and the processes to manufacture and test the cars work from very detailed specifications, and they can refer to a long history of what worked, what didn't, and why. Cars got incrementally better during my lifetime along multiple axes: safety, efficiency, performance, features, comfort. And the engineers who design and build cars can predict costs, schedule, and promise a working vehicle. It doesn't always go right, of course, but real engineering processes like that are a couple of orders of magnitude better at delivering as promised than software engineering.

Literally decades of study have gone into making programming a more engineering-like craft. A lot of good ideas got written about and tried, but then forgotten and rediscovered and renamed, repeated over and over. When companies stopped mentoring and expected junior programmers to hit the ground running, trained at university or a boot camp or on their own, they discarded the apprenticeship part of the training actual engineers and people who work in the skilled trades still get. It doesn't help that juniors start calling themselves senior when they get their third or fourth tutorial project to run, but I think the failure to cultivate and grow talented programmers goes beyond a few eager young people wanting more pay. I feel lucky that I got into programming at a time when employers had to put some effort into training and mentoring, and growing their people into more responsible roles, because I don't see that at all today. That really was better in the past than it is now. And I think that's what separates real engineering from just calling something engineering: a shared body of history and understanding of the fundamentals of the craft, what has worked and what hasn't, and some provably correct principles that lead to predictable and repeatable software development.

If you want to learn about how people get training in a difficult and competitive trade, learn their craft through apprenticeship and mentoring, read how Anthony Bourdain or Gordon Ramsey worked their way up as chefs. They didn't just get out of culinary school and call themselves masters the first time they didn't burn scallops. Mastering any complex skill takes time, practice, and the opportunity to learn from experts, and to learn from your own mistakes and the mistakes other people made.