Treating data/variants.tsv.zip as source data:

1. Write a Kafka producer which publishes records onto a Kafka topic
2. Come with a strategy to populate published records into an Elasticsearch index
3. Come with a strategy to populate published records into a postgres db table [Optional]

We're trying to mimic a 1 source -> 2 sink scenario, but you can choose to treat #3 as optional. Please craft solutions prioritizing maintainability and best practices.

• A docker-compose setup is provided as a starting point, which sets up Kafka, Zookeeper, Elasticsearch, Kibana & Postgres
• Feel free to use a more compact setup which reduces # of individual pieces (eg: lenses.io)

Python or a JVM language (Kotlin, Scala, Java) is preferred, but feel free to employ any programming language