How creating ETLs is very easy, for most of the companies it is what will fulfill their needs.
The reason of doing this: Many companies ask me if I know Informatica or SSIS, I tell them, I have been creating ETLs for more than 9 years now using programing languages like Perl, R, PowerShell. My ETLs work quite well, they are fast and eficient. I think the question they ask is like to ask an accountant if they have used LibreOffice Calc
Here I am using R as an example. Why R?: R can be used mainly as an statistical language, but to do that it needs to be good at data handling and it is good. Specializing in R will also help you to see all the advanced analytics things you can do with it. Also you will see some of functional programming. R has a very big community. I suggest you to read this link: https://www.reddit.com/r/Rlanguage/comments/1aiz4kj/why_r/?rdt=48701
For databases I will be using SQLite with R package RSQLite(https://cran.r-project.org/web/packages/RSQLite/vignettes/RSQLite.html) and Apache Derby(https://db.apache.org/derby/) with JDBC driver to show two different databases and two different methods to access a database. Although Apache Derby allows to have a client/server mode (which I have used for concurrency), I will only use the embeded driver here.
Here a link to download JDBC drivers for what they call common databases: https://www.codejava.net/java-se/jdbc/jdbc-driver-library-download