They're transitioning right now.
They just got major Series E funding, and different factions are fighting over whether it's time to write enterprise software or if it's a startup with too much money.
Top-level tech decisions are driven by an old-school perl hacker who micromanages and wastes his time dictating trivia. Some managers want to fix the existing broken code (someday, when there's time). Others just want to focus on cranking out new features as fast as possible. Both types prefer "tried and true" technology: much of it was deprecated 10 years ago.
They spent 2 years getting a microservice into production. They totally missed the points that could have made this a good move.
But they're driving hard to force devs to write unit tests, because those will fix everything.
Data scientists write a lot of the problematic code. Then they hand it to "real" developers to convert from homework project to something serious. Usually without any explanation of how it works or what it's supposed to do.
They're trying to improve. They hired a new architect in February. He gave up and moved on in September.
Frequent free meals, well-stocked kitchen. Unlimited PTO
Good luck finding time to take any PTO