How to improve speed of R code by replacing complex and slow plyr steps with data.table or dplyr?

I've been learning R by searching how others have done things on Stackoverflow and as a result, I've become acquainted with plyr syntax. I have the following four plyr calls with ddply that are the rate limiting steps to my code. My data approaches the hundreds of thousands to millions of records and my code mostly flies thanks to data.table and is only limited by four rate limiting but critical plyr steps. I would like to replace these with dplyr or data.table, but I've been struggling to replicate the syntax and would appreciate any help. I've been learning R by searching how others ha