My understanding is ML engineer gets machine learning models running in production. They would own the infrastructure that enables teams to deploy ML models, in

My understanding is ML engineer gets machine learning models running in production. They would own the infrastructure that enables teams to deploy ML models, interface with engineering teams to agree on contracts/endpoints/UI/etc, and work on scaling models when/if that time comes. I would view Data Scientist and ML engineer as distinct sister roles, and not as a subset. I've worked in environments where ML engineers were "data science plus", and those were really unhealthy work environments - it ended up feeling like a prestige war with lots of cookie licking and turfing. When you're at the point of needing differentiated roles, you really need someone to focus on infrastructure to deploy ML models and someone else to focus on managing stakeholders, wrangling data, and developing proof of concepts.