In /usr/local/bin, /bin, /sbin, and perhaps /opt or ~/github are a variety of programs somebody else has written that you can run.
For example cut is useful in data "science". [datascienceatthecommandline.com] <data_file_of_interest.csv cut -d, -f3 will pull out the third column. (Hopefully—watch out for those "!) Another program (awk or expr) can then sum the column — or maybe r -pie 'as.numeric(stem(stdin()))' (littler) will create a histogram for you. GNU parallel (pi.dk/3) does a lot of these things too.
Various of these tools are written in C by someone who thought about it long enough to get it right. Not that the program necessarily does what you want, but the way it parses and allocates memory is not horrible.
A few programs I mentioned above (awk, littler, and parallel) are more like meta-programs — they’re programs that write programs. See the awk book (archive.org).