@Owenvii made a post over at (https://devrant.com/rants/2359774/...) and I want to write a proper response.

The biggest thing you have to look out for as a new dev is the jobs which you accept to begin with.
This isn't minimum wage no more, this is "big league", well, maybe not apple or google big league, but it's not $9.25 an hour either.

Basically you don't want to work anywhere where 1. your labor will be treated as a highly disposable commodity. 2. where the hiring manager doesn't know how to do the job themselves.

The best thing you can do is, if you're new, and just breaking through (and even if you're not), is ask them common questions and problems/solutions that crop up doing the work. If they can answer intelligently that tells you the company values competence (maybe), enough to put someone in place who will know ability from bullshit, merit from mediocrity, and who understands the process of progressing from junior dev to a more involved role.

It also means they are incentivized to hire people who know what they're doing because the training cost of new hires is lowered when they hire people who are actually competent or capable of learning.

Remember, an interview isn't just them learning about you, it's your opportunity to interview *them* and boy, you'll be making a BIG mistake if you don't.

Ideally you want them to ask you to pair program a problem. If your solution is better than theirs then they aren't sending their best to do interviews, and it tells you the company doesn't fire incompetents. The interviewers response can tell you a lot too, if they critique your work, or suggest improvements, and especially if they explain their thinking, that is an amazing response to look for, it says the company values mentorship and *actual* teamwork (not the corporate lingo-bingo 'teamwork' that we sometimes see idolized on posters like so much common dogma).

Most importantly, get them to talk about their work and their team. If they're a professional, it'll be really difficult to pry anything negative about their co-workers out of them, but if they're loose-lipped and gossipy thats a VERY bad sign, regardless of what they have to say.

Ask to take a tour and do a meet n' greet of who you will be working with. If they say no, then it's no thank you to a job offer. You want to take every opportunity to get to know everyone there, everyone you'll be working with, as much as possible--because you'll be spending a LOT of time with these people and you want to rule out any place that employs 'unfireable' toxic assholes, sociopath executives, manipulative ladder climbing narcissists, and vicious misery-loving psychopathic coworkers as quick as possible. This isn't just one warning flag to look out for, it's the essential one. You're looking for the proper *workplace culture*, not the cheesy startup phrase of "workplace culture", but the actual attitudes of the team and the interpersonal dynamics.

Life is really short, and a heart attack at 25 from dipshit coworkers and workplace grief can and will destroy your health, if not your sanity, the older you get.

Trust and believe me when I say no paycheck is too grand to deal with some useless, smarmy, manipulative, or borderline motherfuckers at work constantly. You'll regret it if you do. Don't do it. Do you fucking do it. Just don't.

Take my words to heart and be weary of easy job offers. I'm not saying don't take a good offer that lands in your lap, I AM saying do some investigating and due diligence or the consequences are on you.

Turns out you can treat a a function mapping parameters to outputs as a product that acts as a *scaling* of continuous inputs to outputs, and that this sits somewhere between neural nets and regression trees.

Well thats what I did, and the MAE (or error) of this works out to about ~0.5%, half a percentage point. Did training and a little validation, but the training set is only 2.5k samples, so it may just be overfitting.

The idea is you have X, y, and z.
z is your parameters. And for every row in y, you have an entry in z. You then try to find a set of z such that the product, multiplied by the value of yi, yields the corresponding value at Xi.

Naturally I gave it the ridiculous name of a 'zcombiner'.

Well, fucking turns out, this beautiful bastard of a paper just dropped in my lap, and its been around since 2020:

https://mimuw.edu.pl/~bojan/papers/...

which does the exact god damn thing.

I mean they did't realize it applies to ML, but its the same fucking math I did.

z is the monoid that finds some identity that creates an isomorphism between all the elements of all the rows of y, and all the elements of all the indexes of X.

And I just got to say it feels good.