Large python backends are an absolute hell to deal with. Bad performance, no static typing, no clue what's going where and difficult to understand the flow even

Large python backends are an absolute hell to deal with. Bad performance, no static typing, no clue what's going where and difficult to understand the flow even on modern IDEs. I guess most people haven't dealt with that and so even though it is old, it is cool. Python for ML/DS use cases is chef-kiss though. This and newbie friendliness is where the love for Python comes from imo