Ask HN: How to get out of Tech and still make a decent living?

Fascinating. I've been interesting in going towards law school, but am not sure I'd actually like to practice law at the end. This uncertainty plus the cost of attending, has been discouraging.

I wonder how she was able to go through both since each comes with a steep cost

Law School: $34,300 (avg per year) * 4 = $137,200

Medical School: $32,889 (avg per year) * 4 = $131,556

137,200 + 131,556 = $268,756

One would need to make quite the pretty penny to recoup those investment costs

Yes!

I'm pretty un-hireable right now. I have 4 years experience as a dev, 1.5 in sales, 1.5 in marketing, and One of the most fun things about my path in the last couple of years has been embracing the actual problem the business is trying to solve, instead of just the neat tech that powers it. For me, it's a lot more fun to work on problems that have that tightened feedback loop (e.g., somebody gives you money because you solved their problem). It meant breaking a lot of the ties in my head between easy/hard and worthless/valuable -- sometimes the most valuable thing about your software is the weekly update email.

To OP, without a BUNCH more information about what you like and what you can do, I can't suggest a new path. But you probably know something about the products and markets at the companies you worked at. See if any of the other departments have openings for somebody with deep product knowledge. It's easier to switch careers within the same company than it is to jump companies to a different role.

My 18 month year old wrote this sorry can't delete it

I hear you, and hope you can find something that works for you. A cousin of mine, an aeronautic engineer, went on to become a patent agent (editor/writer/evaluator) at a big patent firm. He needed to do some legal training, but it was less than a year -- not a full law degree. He couldn't be happier - he is paid to read and be up-to-date on a lot of interesting stuff, has conversations with interesting people all day and (at least in his firm), has a job that is as 9 to 5 as he wants it to be, no pressure, no crunch time, etc.

The down side is that patents are evil.

> And no respect for experience, unless you demonstrate on a whiteboard with your implementation of a sorted bubble tree list.

I will be a dissenter here: I have interviewed people with 20 years of experience, and who have shipped products, who had gross misunderstanding about the tech stacks they are using -- to the point of taking multiple (2x - 10x) times of both implementation and resources than a reasonable solution -- just reasonable, not compared to an "optimized" (for resources) or "quick and dirty throwaway" (for time) solution. And most of the times, these people were oblivious to their (lack) of knowledge or understanding.

Your post reminded me of a specific interview, someone with 20 years of experience, who's answer to a question (essentially, a database join, which could have been done inside the select but wasn't), which was an O(n^3). Would have worked reasonably well even on an n=1000 table, but our table had n=10,000,000, and this was specified in the requirements. After grilling him a bit about runtime estimates (to which his basic answer "who cares? computers are fast enough nowadays"), he did acknowledge that as written it would take forever, but then added "But it doesn't matter, the compiler will optimize this anyway to the best possible O(n) solution, and it will be fast". This was about a C loop, in 2005.

I thanked they guy and declined his job application. I am sure his takeaway was that I was a snotty employer who only wants sorted bubble tree lists and arcane academic stuff, and that I couldn't appreciate him shipping products for 20 years. But had I taken him on our team, I am quite sure it would have been a failure: At that place, we had a successful fire-and-almost-forget shipping culture, which required doing back-of-the-envelope runtime estimates (among other things) to make sure things actually satisfy requirements -- and his attitude did not fit.

Another similar "20 year experience" story from a friend of mine who worked on a system running a moderately sized financial exchange: In small scale tests everything worked perfectly well, but once real load testing started, it was clear that it can sustain less than 10% of the existing load. (This was a 2nd-generation system, which was to take over existing trading). All measurements pointed to hard drives being the culprit, and a $50,000 SSD (whopping 10GB capacity, IIRC - this was 2002) was bought, and was able to just get the existing load working, with very little room for increased capacity.

At this point, my friend was assigned to review that the solution was properly implemented, and he discovered to his horror, that one of the 20-year-experience authors was unable to figure out how to do a "sprintf()" with unknown-in-advance result size, so he instead created a temporary file, used "fprintf" into it, allocated the memory according to the resulting size, read the data, and deleted the file. Changing this to an snprintf-and-if-failed-reallocate-buffer-and-retry scheme sped the system up some 20x (compared to the SSD!), and made the SSD redundant. The original guy's response when shown the solution was IIRC, basically a dismissive "oh, yeah - I guess I should have known about snprintf, it wasn't there when I learned C, of course the SSD was the right solution given the constraints".

I'm not saying experience is useless, but it is an appeal to authority, which most of us (especially here on HN) dislike when we are expected to revere it. If one can't show competence without referring to (vague, abstract, generally unverifiable[0]) experience, then perhaps one is not as qualified for a job as one thinks.

Just to clarify: This is not personally directed at OP - in fact, the vibe I get is that OP can easily pass sorted bubble tree list interviews and just got tired of it.

However, I'm just pointing out that anecdatally, I've seen too many "experienced" developers who are oblivious to the costs/benefits of employing them, to the point that I completely disagree with the prevailing (even on HN) sentiment that experienced developers shouldn't need to do the kind of interviews that junior devs do.

[0] Being a member of a team that did well is not, on its own, proof of anything. the dailywtf is full of "Paula"s and others who have spend months -- even years -- on teams that shipped, and the real world is too.

Especially if you are not going to practice at the end, consider going abroad to study (might be possible even with family, definitely possible without).

The costs in western europe are about 15%-20%. The costs in eastern europe are half that, and in some places in Asia it is cheaper still. If you want to practice in the US, you will have to take some local-law courses (law) and pass the bar/board exams - but it will still be infinitely cheaper than in the US.

And it will be an adventure.