Very Nice !!

few remarks on the article :

1. kafka-consumer deployment is "0" replicas

2. link to keda/kafka is broken , the correct link is

https://keda.sh/docs/1.4/scalers/apache-kafka/

3. you need to first do

eval $(minikube -p minikube docker-env)

otherwise the minikube will not have access to docker deamon

4. in kafka-scaledobject.yaml you need to use the keda namesapce not the default namesapce

5. I dont see a code for commit the consumer offset in the Consumer code

6. if you create a topic with single partition there is not much meaning to scaling here.

7. you always send to partition "0" -> again , this is not scalable and there is no much sense in scaling more pods since only 1 pod will be able to read from it.

8. well done for choosing Vertx ! clean + simple.