After being screened by a questionnaire to get to this stage, the first interview was behavioral and in a speed dating format. We were brought up in groups of about a dozen applicants at a time, and there were a lot different numbered tables with two DL people at each one. They had us rotate through, sitting at each table for a few minutes and answering 1-2 questions, and if there was time you'd get to ask questions, and then a bell would ding and you'd switch. The questions were standard interview things like "Tell me about a time you had to deal with conflict," with a few silly icebreaker ones like "What would you want your super power to be?" It was a little nerve wracking, just because of the format and having so much chaotic activity going on around me, but I did well because I'd run through typical questions in advance so felt prepared. After this there were three technical interviews. The first two were pair programming with a partner (a randomly assigned other applicant) to solve fairly simple problems (write a function to do a simple math things, string manipulation, etc.). We had a llimited amount of time, and there was an extra problem if we got through the first one (I didn't have time to tackle it but look at it and tried to solve it on my own later, which was smart because I got it again in the next technical). We used something like replit, so could run and test the code while working, and because it was an apprenticeship position weren't expected to know anything about JavaScript in advance, but were encouraged to google to try to figure it out and/or write pseudocode. Though since they did notify us in advance that the technicals would be in JavaScript, we did have the opportunity to research a bit. Labs people would walk around the room and watch over our shoulders, I'm sure observing communication skills and how we approached problem solving cooperatively, and at the end we were to email our code to an address provided. The second paired exercise we did remotely (due to pandemic), with a different person but same idea. The third technical (also remote) was done independently and was a little more complicated but nothing crazy. You were encouraged to read up more on JavaScript in the interim, so it seemed fair that it would be a bit involved. I don't think they expected everyone would necessarily finish with working, flawless code and I believe pseudocode was acceptable if the thought process was sound. I know at least one other person who was accepted didn't finish all the challenges with working code, though I think I did and that probably helped me.