A poor place to grow, an easy life if you're prepared to stagnate.
Pros
The work was quite easy, and there was little pressure to deliver.
Cons
The data science team has a chronic lack of experience, and is lead by people without technical expertise. The result is that projects are haphazardly slapped together without any real thought given to use-case, meaning projects drag on for years and often lead to marginal or no impact on the business. Even worse, the lack of technical expertise means that no thought is given to appropriate solution architecture (e.g. projects are "productionised" through shared folders and Excel documents, low-code RPA processes are used instead of actual software), so the often pointless projects also come saddled with a mountain of technical debt, with little appetite from management to fix it, or invest in the people and skills needed to do things properly choosing instead to endlessly rotate graduates and apprentices through the team). There was no code review process, the team had not adopted minimum engineering standards such as appropriate testing or correct use of version control. Appropriate tools are not used for development, because management dictate the solution design, despite lacking appropriate experience. ML engineers are expected to be code monkeys implementing management's poorly-conceived ideas; it is not a good place to learn new skills or develop meaningful experience as an engineer, because there are no experienced people to learn from or work with, and the projects rarely add meaningful value.