Worked on the apache/nifi repository to enhance the reliability and deduplication of the Kafka Consume Processor, focusing on improving data integrity and resource stability in distributed streaming pipelines. Addressed a critical bug by ensuring Kafka consumers are not reused until their sessions are fully committed, which prevents duplicate record consumption during restarts. Implemented robust error handling to manage session rollbacks in the event of exceptions and enforced concurrency limits by capping active consumers to the configured maximum. Utilized Java and leveraged expertise in Kafka, NiFi, and testing to deliver changes that improved the maintainability and stability of the ingestion path.