I've found the opposite. It can take a long time to get your head into that space but it produces much more maintainable and robust systems in my experience.
I have a similar experience. I've worked with a product that consists of tens of thousands of stored procedures implementing all business logic. Because you are right next to the data you can get away with doing many thing you would have not otherwise.
Also it acted like a barrier, the core team who understood the business well worked on the stored procedures, keeping things fast and stable.
Another team was doing the tech to interface with the database and UX/UI was then implemented in various technologies in parallel and changed/got recycled every few years to whatever was in fashion.
The difference in culture between the two teams was huge, it seemed to work out well in their case.
It also saves a lot of extra work. Often with more conventional systems you have to pull the data out. Move it into a structure native to the language doing the processing.
In the database you don't need to do any of this. The data is there. The type information is there.
