I've found this tutorial -
http://artsy.github.io/blog/2013/02/15/infinite-scroll-with-mongodb/
but I cannot seem to grasp what parameters they are sending and why?
Pagination and infinite scrolling are best implemented with range queries, which the blog post you cited demonstrates. Skip and limit are conceptually simple, and are actually quite fine if we're only dealing with a few pages, but performance starts to fall off drastically as the page counts increase into the hundreds.
When you issue a query (hopefully indexed) with a skip, MongoDB needs to go to the first result and walk forward "skip" number of times. For indexed queries, this means walking the index, which may not be entirely in memory. Contrast this to a range query that does not use skip, where our criteria requests data greater or less than that which we've last seen, MongoDB is able to jump right to the first result and we need only iterate through "limit" documents.
The following answer and blog posts explain the subject in more detail: