Are companies lying about STEM jobs?

To wit: in Go I just benchmarked the standard library’s doubly linked list’s “PushBack” method vs the builtin “append” on slices (an array-backed list type) and append was twice as fast despite the theoretically correct answer that Linked Lists are faster to insert into. (50ns/op vs 106ns/op)

Medicine. Finance. Lots of business type jobs.
I’m only taking about bachelor’s level - get a Ph.D. in CS or the right types of engineering and you’re in good shape.

That’s a bullshit question without putting it into the context of a given application. This is the type of loaded question where the interviewer wants to prove they know more, that they are superior to the interviewee. Yeah, we’re glad you don’t work for a big company anymore too. I guess garbage collection does work, kinda.

Wait, what? As has already been pointed out in this thread, that’s data structures 101 stuff. Nobody who doesn’t know the definition of “linked list” and “array” and can’t tell you why they’re different is ready for even an entry level software development job. No context is necessary for a good answer; I’m struggling to think of a circumstance where context would even change my answer. Stuff only gets harder from there. I agree that you shouldn’t be WRITING linked lists any more, but you still have to know what one is and the computational cost differences between it and arrays.

Seriously, this is like a person applying for an English language instructor position who doesn’t know the difference between a noun and a verb. If an interviewee can’t answer that question (and it isn’t just a misunderstanding of some sort), I am superior to them, at least as far as ability to do the job goes, and they’re not going to get hired. Sure, we can train in-house, but that’s basically starting from scratch. It will have been covered in their first course or program, and reinforced continually ever since.

I think a more telling question is usually a list (array-backed or linked, doesn’t matter) vs a fast hashmap/dict type for a given task. It’s a surprisingly difficult design issue in a lot of cases and I think you can find out a lot about a person’s reasoning about certain tasks based on which they’d pick. It’s the kind of question where the best answer could even be “benchmark it, then decide” or “which is more maintainable for the task?” They have to consider things like “how often do we expect having to search for a specific element?” It often doesn’t have “nice” clearly expected theoretical answers like the linked vs array list question, especially since those “nice” answers can be wrong in real world cases.

A bunch of STEM nerds have hijacked this thread.

The original statement was, “What are the advantages and disadvantages of using linked lists vs. arrays?”

Not

"know the definition of “linked list” and “array”

There’s a huge difference. And where did this come from: “I agree that you shouldn’t be WRITING linked lists any more”

No longer a need to write a doubley linked list of pointers to arrays? Really? And no, you aren’t going to know the computational cost differences between it (linked lists) and arrays until you couch it terms of an application.

Jeezus. Aren’t there any swinging dicks in this thread?

The blogger Chemjobber has a series of articles posted over time with the contention that STEM is really TE. You only hear about computer programmers and various engineers when it comes to STEM discussion in the media.

They must not have enough work to do… ;):stuck_out_tongue:

Whether the workers are Indian or “foreign” (do you mean foreign-born? foreign-sounding? something else?) and whether they are H-1B workers are two different issues. H-1B workers are allowed a max of six years in H-1B status, after which they must have reached a particular point in the permanent residency process in order to keep extending their H-1B status, switch to another visa category, become permanent residents in some other way (via marriage, etc.) or leave the U.S. H-1B status isn’t indefinite.

How do you know that a chunk of these “foreign” employees aren’t permanent residents or U.S. citizens? I’ve got all kinds of clients who started off as H-1B workers, but are now permanent residents and/or citizens and are sponsoring family members now. And they’re still working at their STEM jobs.

Eva Luna, U.S. Immigration Paralegal

Yikes. Why the recreational outrage? I’ve been on both sides of lots of interviews and i’d consider it a softball question. The question I usually ask is a lot harder. High-end companies like Google are harder still.

As for the OP my experience is completely in the CS/SE field and there is definitely a shortage of good candidates. It takes my company months to find a match at junior and senior levels.

Agreed. Here’s a Google interview question I got right off the bat. Of course, if you spent all of your time in school farting around with HTML5 instead of in a hex editor examining stack traces, you’d never get it.

Question: Explain the significance of “dead beef”?

Source: here.

None of the questions in that article have ever been Google interview questions.

One possibility:
The employer gets an employee who is not likely to be looking for better work and is unlikely to balk at either unreasonable or unethical demands. The H1-B worker who seeks a better paying job or or a job with better working conditions is liable to have their particular H1-B pulled, leaving them with nothing but a trip back to their native land.

I do not claim that this is the prevalent situation. I know that it happens. I worked with a manager who explicitly talked of hiring H1-B applicants for that reason. I also worked with a man who was hired as a temp when an H1-B candidate got caught up in paperwork and who was released as soon as the H1-B arrived, even though he was doing exactly the work for which the H1-B was supposedly uniquely qualified.

The US government spends a lot of money educating foreign nationals. My PhD was funded almost entirely by the taxpayers. Outside the one year I taught, that’s 3 years of tuition, 4 years of stipend and health benefits, plus supplies and overhead/indirect. I don’t know what my fully loaded cost was, and I don’t know how much the US spends on non-citizens. Many scientists also go on to work as government-funded postdocs.

I’ve seen, time and again, the frantic hoop-jumping these folks go through to stay in the US. Many of them ended up going back home. I feel like we should try to keep them if we paid that much money. If we don’t have the room, then we shouldn’t be paying for their education.

Strictly speaking we were probably paying for your research, with your education being a side benefit. I just wish a higher percentage of the visas were reserved for people like you and fewer for easy to hire programmers.

I don’t recall which was greater, tuition or stipend, but yes, the PI applied for money to fund the research. Conducting that research was my education. The whole system was kind of weird. No student ever actually paid that tuition.

Tuition is funny money. I don’t know about you, but when I got my PhD I didn’t go to a lot of classes except a few seminars.

They sound like typical management consulting case interview questions. “How many flights land at JFK every day?”, brain teasers like the “Monte Hall” problem. I can’t speak for Google, but some companies do ask those sort of questions.

I can’t speak for all of Google but those are definitely not software engineering questions.