Did you read the article?

  > It’s got 25 public methods. Two of those return further interfaces that
  > have a combined total of 12 public methods. That’s a total of 37 public
  > methods for which you have to build and test in an implementation.

I don't think implementing 37 methods in 30 minutes is particularly reasonable.

I have. Pretty much you already have to have the whole implementation in your mind before you write a single line of code. Shouldn't be an issue since you should cover that exact implementation in your uni algorithms classes. Well I've worked with Java since 07, so perhaps writing the implementation its a bit more intuitive for me. I do agree that timing is tight so writing unit tests wouldn't even a consideration.

And do you have the whole implementation on your mind all the time?

If not, does the time spent thinking and researching it before not count as implementation time?

At my last interview I had this entire pdf in my head:

www.ee.ryerson.ca/~kclowes/documents/eaads.pdf

No I wouldn't say it counted as implementation time. Same way as driving to work doesn't count as work (as unfortunate as that is).

Linked lists are certainly one of the easiest structures. For me, it's strictly a time thing here. An hour? Maybe not needing List<T>? With one of these, doable, though I'd still argue it's a bad interview. But the short time plus extra requirement seems really unreasonable.

I can see how the fact that its an interview will make it significantly more stressful to fit in the required time. Rating your reaction was probably another part of the interview altogether. The only real recommendation I can give is - have lots of coding practice before you go. Hackerrank, codewars sites are pretty good for that.