Real-Time Data Pipeline: Kafka to ClickHouse with Python

Build a production-grade real-time data pipeline using Python, Kafka, and ClickHouse. Features async processing for thousands of events/second, batch inserts optimized for ClickHouse, auto-retry with exponential backoff, and built-in metrics. Tested at 500 events/second with <50ms processing time. Use cases include user behavior analytics, IoT processing, and financial monitoring. Full implementation includes 500+ lines.