the article is long overdue. so many times i have seen data scientists glue together non production ready code and libraries expecting someone else to finish t

the article is long overdue. so many times i have seen data scientists glue together non production ready code and libraries expecting someone else to finish the job and put the project to production. in various places i worker at they have soon realised that we mostly and exclusively needed data engineers - glorified DevOps people that make production ready data pipelines - more than we needed data scientists to do “modelling”