Study Confluent CCDAK Streams and Connect: key concepts, common traps, and exam decision cues.
On this page
CCDAK is still a Kafka developer exam, but Confluent’s public role description includes platform capabilities too. That means you need enough awareness to know when a problem belongs to Streams, Connect, or Schema Registry instead of to the plain producer or consumer APIs.
Platform-boundary chooser
If the question is mostly about…
Strongest first focus
stream-processing application logic on Kafka data
Kafka Streams
source or sink integration runtime
Connect
schema management and compatibility
Schema Registry
plain producer or consumer API behavior
Kafka core client path
What the exam is really testing
If the scenario shows…
Strong reading
app built with Streams API
stream-processing boundary is under test
connector runtime or integration issue
Connect boundary is under test
serialization and schema evolution issue
Schema Registry path matters
basic send or poll behavior
Kafka core client logic matters most
Common traps
Trap
Better rule
assuming every event-streaming problem is a core broker or client problem
identify the right platform boundary first
using Streams language when the scenario is about connectors
processing API and integration runtime are different layers
forgetting that Confluent platform capabilities still depend on Kafka fundamentals
core client and partition behavior still underpins the platform
Decision order that usually wins
Decide whether the issue belongs to Streams, Connect, Schema Registry, or plain Kafka client logic before tuning anything.
If the stem is about application logic over Kafka records, stay in the Streams boundary first.
If the stem is about source or sink integration runtime, move into the Connect boundary first.
If serialization or contract evolution dominates the stem, shift into the Schema Registry lane before broker tuning.