Kafka Producer and Consumer with Schema Registry for an Event-Driven Service

Problem/Feature Description

The backend team at LogiTrack is building an event-driven parcel-tracking service. Parcel scan events are produced to a Kafka topic called parcel-scans, serialized using Avro, and consumed by multiple downstream microservices. The team decided to use code-generated Avro classes (from .avsc schema files) rather than generic records so that each service has compile-time type safety. A senior engineer has asked a team member to produce a reference Java project structure — with pom.xml, producer code, and consumer code — so that new services can onboard quickly and consistently.

The parcel-scans topic sends events where both the key (parcel ID string) and value (scan event) are Avro-serialized through Schema Registry. Multiple event types are expected in the future on the same topic, and the team wants a schema organization strategy that will scale gracefully as new event types are added without tight coupling to topic names. For now, they want a reference implementation that can be reviewed and adapted. The producer and consumer code does not need to run against a real cluster — it should demonstrate correct configuration and usage patterns.

Output Specification

Produce the following files:

• pom.xml — A Maven POM showing correct dependencies and repository configuration for Avro Schema Registry serialization. Include the Avro Maven plugin configuration for code generation from .avsc files.
• ParcelScanEvent.avsc — An Avro schema file defining a ParcelScanEvent record in the com.logitrack.events namespace, with fields: parcel_id (string), location (string), timestamp (long), and an optional notes field (nullable string).
• ProducerConfig.java — A Java class (no main method required) that builds a Properties object for a Kafka producer that uses Avro serialization with Schema Registry. Choose a subject naming strategy appropriate for the multi-event-type scenario. Include all required and relevant configuration properties as comments or code.
• ConsumerConfig.java — A Java class (no main method required) that builds a Properties object for a Kafka consumer. The consumer should be configured to deserialize into the generated ParcelScanEvent type-safe class rather than a generic untyped representation. Include all required configuration properties.
• design_notes.md — A brief explanation of the subject naming strategy chosen and why it suits the multi-event-type scenario described above.