Back to Blog
Real-timeEvent-drivenMachine LearningAI Workloads

Real-Time for Everyone: Building a Zero-ETL, Event-Driven Pipeline for Nairobi’s Bike-Taxis

8 min read  · 1,491 wordsBy Orandi Felix

EMQX’s rule engine is a game-changer: you can route messages to Kafka, HTTP endpoints, or databases without writing custom consumers.

ClickHouse’s `argMax(rider_id, ts)` picks the most recent `rider_id` for each aggregation window—a common need in event streams.

These numbers are for a **single node**. Horizontal scaling would require adding load balancers and replication.

Share this article: