While SQL has become the lingua franca of analytics engineers and is widely used by other data practitioners for modeling and analysis, it can be ill-suited for composing linear pipelines of transformations. In contrast, PRQL, an interpreted relational query language recently OSS’ed by Maxmillian Roos, is explicitly designed to transform data with abstractions like variables and functions (which help users avoid repetition and write cleaner code). With PRQL, which transpiles to SQL, each line of code represents a transformation of the previous line’s results.