HashSet<T> stores unique values with O(1) average-case membership tests. It's backed by a HashMap<T, ()>. I use it for deduplication, membership checks, and set operations (.union(), .intersection()). Common methods: .insert(v) adds, .contains(&v) checks, .remove(&v) deletes. For iteration, .iter() yields &T. Sets are unordered; use BTreeSet if you need ordering. HashSet is great for "have I seen this before?" checks in parsers, dedup in data pipelines, and implementing set-based algorithms. Like HashMap, it uses a randomized hasher for security. The API is simple and efficient for most set use cases.