Arroyo is a stream processing engine; users write SQL to build real-time data pipelines. Now SQL is a great language for data processing, supports a huge variety of functions, and is adaptable to a wide variety of use cases including streaming data. But as powerful as SQL is, sometimes there's a computation that just isn't easy (or possible) to express as a SQL expression. Maybe you need to parse a custom binary data format, implement a complex aggregation strategy, or call some pre-existing bus...