After deciding to eliminate the use of a previously used statistical programming language, NPD wanted to standardize its processes on R and Python. As the company’s Research Science team already had a third of their code in R, they focused specifically on standardizing on the R language.
In preparation for this move, they decided to offer training to ensure the team was comfortable with packages like dplyr and with interacting with Spark from R using sparklyr. NPD had a wide mix of folks interested in the training: some who had never programmed in R and some who had a lot of experience with R, but not with packages like dplyr.
They considered several training options, but were not happy with any of them. Many training options were just videos, and watching someone else code didn’t give their team the hands-on experience they needed. Some courses offered limited interaction, but certain topics were missing and others weren’t exactly relevant to their needs. This would have led the team to have to cobble together a couple of courses and/or supplement a course with their own content.