The currency of the modern microservice architecture is real-time data, allowing applications to make informed decisions on the most recent relevant data our organization has to offer. Layer AI models into the mix and the demand for fast, reliable access to quality data grows exponentially. Apache KafkaⓇ has emerged as the de facto standard for event-driven architectures. But Kafka isn’t just some “dumb pipe” messaging system. Let’s take a developer’s level view at this immutable, partitioned, distributed, durable log.

Starting from the protocol level, we’ll unpack the key terminology of Kafka’s internals - topics, brokers, producers, consumers, and offsets. Then we’ll go a level deeper to demystify how a Kafka cluster delivers on the promises that power functionality at 80% of the Fortune 500. We’ll step beyond the slides to look at some Kafka client code, then end our discussion by looking at other parts of the Kafka ecosystem for event processing and connecting to external systems.

Whether you’re a seasoned Kafka professional or just Kafka-curious, there’s something for you in this talk to further your journey in building event-driven applications.