Stream processing is used for both service integration and large scale data processing in Apache Kafka, Spark and, Flink. Neo4j integrates with all these libraries.
If you run into a bug, rather than posting it here, please create a GitHub Issue (Spark), (Kafka).