While I agree you that the "no-code" platforms are not the demise of the data scientist, I find the biggest reason for this is missing in your article. Which is the fact that finding and graphing the data is the smallest part of a data scientist's job.

The complexities of making data useful fall in understanding the data itself, just as the complexities of being a doctor are not found in the ability to read a blood test or MRI, but in knowing what to do with the information found.

Plotting a chart in Python (or Spotfire, as a no-code example) is easy - as you say. Anyone can do it. But knowing how to munge the data set, include/exclude outliers, determine cause from correlation, correctly (not only) identify treands but make use of what the trends are saying, as well as correctly and understandably communicate these insights to people with less familiaity with either the data domain or data science ... these are the qualifying skills of the data scientist.

TL;DR

Graphs without the wisdom and training to glean meaning from the graphs are useless :)