Advanced · 4–5 weeks
Real-Time Order Processing with Kafka
Order service publishes OrderCreated events. Inventory, payment, and notification services consume and react. Saga-style flow with compensating actions on payment failure.
Why this project
Event-driven architecture is essential for senior roles. You can discuss ordering, idempotency, dead-letter queues, and distributed workflows.
Tech stack
Java 17Spring Boot 3Spring KafkaPostgreSQLRedisDocker Compose
Skills demonstrated
- Event-driven design
- Saga pattern
- Kafka consumers
- Compensating transactions
Interview talking points
- Choreography vs orchestration
- Kafka ordering guarantees per partition key
- Exactly-once semantics reality
- Saga compensation example walkthrough
Problem statement
Modern e-commerce order flows involve multiple systems (inventory, payment, notifications). Doing them synchronously in one service creates tight coupling and cascading failures. Event-driven sagas solve this by making each step independent and recoverable.