Well, since all of game dev is problem solving, I'd like to say it all befell upon me on one capacity or another. (: But since that's a kind of a copout answer, I'll focus on some spesifics.
I was new to GDevelop, so in that way there was new things to learn around every corner. Luckily the basics were not too hard to pick up and there were tutorials for everything. The extensions were also a hoot and helped to get some boost to creativity and not have to program everything yourself. I did face many situations where it seemed harder to do some basic sounding things "the GDevelop way", that would have been fairly easy to code in a code based engine.
One thing I had to solve, was having a "load next scene" funcionality. Had to crack open the ol' javascript API and dig around there to get the index of the current scene and add +1 to that to get a reference to the next scene and use that to get the name of the scene to load instead of going through every scene individually to change the name of the scene to be loaded. Since we didn't use version control, this was what I could think of to keep the project a bit more organized among other things. It's funny such functionality is not provided (as far as I know), but there you go.
Things we had to scrap: there were several more mechanics that we were thinking about. Including dashing to 4 directions, crashing through objects, wind puzzles, throwing the package like a bomb... Also there was art that was made, but didn't make the final build. It all sounded possible in the context of 9 days, but then again having had some personnel issues in the first half of the jam, we were quite happy with what we were able to manage in spite of all that. :) There were also a couple more levels, but since we didn't have time to balance everything out, we focused on a 'core' bunch of levels that kind of introduce the player to all the mechanics.
Overall I had alot of fun, and it's not over yet! See yall in the comments sections while giving out the remaining ratings!