Developed a real-time sentiment analysis pipeline for the drshahizan/HPDP repository, enabling live sentiment classification on streaming data. The solution leveraged Kafka for stream ingestion, Spark for distributed processing, and a convolutional neural network for text classification, with results indexed in Elasticsearch for efficient search and retrieval. Kibana dashboards and documentation were added to support observability and monitoring, providing clear visibility into pipeline metrics. The work included updating the repository README with a comprehensive design overview and usage instructions. Implementation utilized Python and YAML, demonstrating depth in data engineering, machine learning, and real-time stream processing within a production-ready environment.