Data Science Is Not Dead
JeanFrancoisPuget 2700028FGP Comments (2) Visits (11390)
Is data science dead really? One can wonder after reading Jeroen ter Heerdt's Data
This is true. Lost of companies now provide apis that rely on some machine learning models internally. IBM is no exception with its Wats
I disagree with Jeroen iwhen he concludes that the world no longer needs data scientists. True, the most common use cases will be, or are already, packaged as ready to use services. But there is a very long tail of use cases where developing a packaged service isn't worth it. We still need data scientists to solve these remaining use cases.
This situation reminds me of operations research (OR). Lots of operations research applications are delivered as prepackaged solutions, be it for supply chain management, price optimization, aircraft scheduling, crew scheduling, production planning, you name it. Yet, many companies still have an OR department (under that name, or as part of their data science department) because they have operations problems that cannot be optimized with off the shelf software of services.