Every day, I hear one candidate ask another, "Have you interviewed with Company X? What were you asked?" This would be a perfectly reasonable question, if it weren't for the fact that they were doing so on a website with thousands of interview questions. Why get 5 questions when you can get hundreds? The truth is that, at most companies, there's no grand system. There's no structure. No one saying, "ok, now, every candidate will get one networking question and it will cover TCP/IP."

There's just... people. Interviewers take a bit of interviewing training, that usually covers oh-so-helpful legalese like "asking candidates about their marital status is illegal", and then off they go! Interviews ask whatever they want to ask. No system, no structure. Just a bunch of people making up their own minds.

Next time you're about to ask someone else what they were asked, stop and think: will this person tell me anything new? Relying on one person's experience for your preparation is much, much worse than relying on the experiences of many.

Instead, do the following:

Check CareerCup.com for programming interview questions, but don't limit yourself to just your company's questions. If you're interviewing for Amazon, be sure to check out the Microsoft Interview Questions and Google Interview Questions. Companies are far more similar than they are different.
Review the questions to get a general feel for what the company likes to ask. If you're doing an Amazon interview, for example, you may notice that Amazon loves object oriented design questions. That'll give you a good idea of where to focus.
Practice! Don't worry about getting the answer to each and every question. Answers won't help you. You need to solve the problems yourself, so you learn the general techniques.

While your questions may vary based on your background, your interviewer, the team, or the prospective company, interview questions are actually more consistent than not. Interviewers don't like coming up with new interview questions for every candidates; it's hard, and results in poorly calibrated feedback.

As far who "invents" the questions, the answer is that it's rarely the interviewer. Employees talk, and questions are shared across a company. When people switch companies, they bring their favored interview questions with them.

I'm still asking my five favorite questions in my mock programming interviews. I don't change them frequently because calibration is more important than creativity. I know how to ask the question, how to lead a candidate to the solution, and exactly how well you did relative to other people. Why change?

Microsoft was famous for its free all-you-can-drink sodas, until Google stepped up the game and offered free lunches (and dinners and breakfasts too). But, as a reader recently noted, do these perks really matter? My personal opinion: yes and no.

No, most perks don't really matter. A free lunch does not fix a bad boss, and bringing your dog to work does not make up for a lack of career growth. Don't be too fooled by these sorts of perks. Instead, ask yourself, what is the dollar amount that this is worth to me? Free lunch, for example, is probably worth just about $2000. Rather than being swayed by some company offering you free lunch, just pretend you are making an additional $2000 in salary.

However, some perks - the less flashy ones - can make a real, substantial difference in your life. Microsoft's "We Pay For Everything" healthcare plan is incredibly important to some people. Or, flexible work hours might be very important to other people. These are the perks that you should really evaluate.

With all that said, while perks may be a flashy recruiting tool, perks are also often a reflection of the culture or history of a company. In Amazon's case, its lack of perks are likely a reflection of its truly being a retail company - not a software company (and of its less profitable history). Is it really fair to judge them by the same standard as Microsoft and Google? However, in other cases, a company's lack of perks might suggest that it doesn't value its employees the way that its competitors do (or that the company isn't as desperate to recruit people).

In the end, no, it generally doesn't matter that much which perks a company offers. You can assign a financial value to each perk by asking yourself, how much would you pay for this perk? Remember that each person values each perk differently, and the dollars you assign might not match the actual price tag. And that's ok.

Assign the financial value of a perk, and then look beyond all that. Ask yourself, why is the company offering these perks - or not? In the end, the culture of a company will make a bigger difference in your happiness than a few thousand* dollars.

* A few thousand dollars can obviously make a difference to some people. I'm assuming here that we're comparing jobs at major tech companies, where the difference between $75k and $77k probably doesn't change your lifestyle. If you're making considerably less or have a unique financial situation, then it can absolutely matter.