I end up working with a lot of data transform pipelines, and Clojure's strong support for names (with namespaced keywords) and generic data structure manipulation makes the code you have to write more succinct.

This kind of task is also extremely well suited to a repl. You define a temporary var, and start threading it through a series of super small functions that only slightly change the data until you get what you want.

    (->> my-data 
         parse
         (filter some-pred)
         (map my-mapper)) ;; and so on, if you need more
        

For me, the big thing is REPL driven development, and immutable data structures. Everything can be inspected. It's easy to get insight into basically any small piece of the program as if you were using a debugger (but you don't have to).