Mmm, I disagree. I do a fair amount of mucking around in notebooks/containers, but I also drop them (when they’re refined enough) into a pipeline where I (not

Mmm, I disagree. I do a fair amount of mucking around in notebooks/containers, but I also drop them (when they’re refined enough) into a pipeline where I (not another person dubbed a DE) can evaluate the integrity of the results on an ongoing basis. Though I do defer pipeline design dilemmas to our architect I do some of the implementation / testing work myself, I firmly feel that for a data scientist to be effective in their role they must also be part Data engineer (and visa versa). It’s hard to compartmentalizations those duties and get results efficiently.