Postmortem "Last Bargain"
Some time passed, voting ended, so it's a perfect time to reflect on the overall experience, analyze what was done right, what was done wrong, and how we can do better next time!
Overall the process for our team was not so difficult like it was the last time we participated, but we decided to find a few new team members - and got support from @sfts & @ri-ra. Sadly, but @ri-ra left before the end, but anyway I think that was the right decision to invite some additional people, it's always fun in the end.
What went wrong
Not a lot, compared to the problems we had during the first participation. The first thing on my list is personal. On the first day, when I was looking for a visual style of the game and was about to prepare assets to build some level I spent about 6 hrs trying to deal with quite complicated tilemap settings, which was not worth it completely. Time-wise it was the biggest fail for me, everyone who has participated in a jam knows, that time is the most important thing. And spending this amount to nothing was an absolute disaster for me personally.

From a code perspective, the biggest challenge was physics. @ramonvicious said that he has never worked with physics-based games before and it was risky, but we underestimated how painful it can be to simulate everything. In the end, we simplified the game's physics and this helped a lot, however, some issues persisted even in the final build.
The biggest pain game-wise from my perspective is the game rhythm and dynamic. We started to have something working from the game's cycle only at the very end of the second day and we didn't manage to polish this aspect. For sure we should make a simplistic prototype much earlier.
What went well
I'll start with the personal one, but, for me, it was quite a move. At the end of the second day, I've checked the project and it was a disaster. It was fractured like a hell and there was no glue to assemble something that we can call 'a game'.
I decided to cut out everything which we were not able to make in a reasonable time and focus only on things that lead to some sort of product. We have dropped physics, simplified controls, agreed on levels structure and content, and started to do only the things critical for the completion. It was harsh, but we managed, and I'm happy that I was brave enough to do the call.

I think we also did grow quite a lot from the previous game jam professional-wise, which helped with the number of things we were able to make.
And finally, thanks to @sfts, we were able to have a nice sound, which wrapped out the whole game in a nice polished way. In the very end, I think we did pretty well. We were able to wrap a nice game out of fractured parts and, what's important, did a few steps forward since the last jam.
You can try it now if you want, we are always open for feedback :) Thanks to everyone who organized, participated, posted entries, commented, etc. It was a nice experience and we'll be happy to participate in the next jam!