I have an interview for a postdoc position next week. The project was running for 9 months or so and they already have some preprints out. This specific position says that the researcher will work closely with the 3 PIs, so it broadens the area for me a little bit since each one has different research areas. Do I need to know in detail the papers that are out in this project? I'm asking this because I've noticed that people say "do your research and look at the latest publications of your future lab".
The advice "look at the latest..." is good advice. Whether it is necessary to do depends entirely on the views of the PI. They may care deeply, or not at all. Certainly it is better to be as prepared as possible, even if their main concerns are elsewhere.
If the position is competitive, then it is likely that other candidates will be as prepared as they can. Due diligence, I guess.
Even if you only bring questions about those papers to the interview you will be better off than being completely unprepared.
In an interview for a postdoc, you should be able to explain two things:
First, what your future research plans are (and why this topic is right for you, due to your past work or other background).
And, second, why you chose to apply for this position, and why you think you are a good fit (or rather, the fit) for the position. This implies that you know what is going on research-wise in this place, and that you can explain how you both will be able to add to the research performed there, and how the topics researched there will help you to form your research profile (first point).
To this end, it is essential that you know what is going on in the place, and what your potential future PI and their group are working on. However, this does not mean that you have to study the most recent papers - it is more important that you have the big picture of where their research is heading. Of course, knowing their most recent papers in more detail does not hurt, but a PI will usually not expect you to impress on specific details of one paper. And it might well be that the last 3 papers, for some reason or the other, don't accurately reflect on their primary research interest. In that case, saying how interesting you find [topic of the paper] could even be counterproductive.
You can flip it around and ask: ‘Is it reasonable to apply to a post doc position without doing some basic research on your future boss and workplace?’ I would think not.
That said, for none of my post doc interviews any detailed knowledge of the papers from the lab was unnecessary. It was much more important to have a big picture idea about the general topic and to be able to follow any contribute to a conversation about that research area.