Tries
A trie is a variant of a -ary tree where characters are stored at each node, and each path down the tree can represent a word:

The nodes represent the fact that all letters leading up to that node in the trie form a valid word-- for example, in this case, MANY
is a valid word.
In interview questions, tries are very often used to store the entire English language for quick prefix lookups. A hash table can do quick lookups to tell us if a word is valid in a certain language, but not for prefixes of a word-- a trie can do this well.
A trie can lookup if a string is a valid word of prefix in time, where is the length of the string (this is the same performance as a hash table, since although lookup is constant time in a hash table, reading all the characters of a string is ).
Remember tries for problems involving lists of valid words.