Separation Between Full Stack Machine Learning Developers and Front-End Full Stack Data Science Models

Is there a needed separation between full stack machine learning developers back-end full stack data science models, deployment and front-side full stack? I think we actually need this separation here, because in my opinion, like when we are talking about a full stack engineer, so this is somebody who mostly focuses on web. So let's say we quickly want to create an application that users can use, like a web application. And this is what they would do. Yeah, so I think there is not a lot of connection between these two, but I might be wrong. But fundamentally the processes, they are similar, but there are some differences as well.