System design interview: when to mention specific technologies vs staying abstract?

there's conflicting advice out there about mentioning specific technologies versus staying abstract in a system design interview. on one hand, saying 'we'd use Kafka here' shows knowledge of real-world tools and their applications. on the other hand, it can derail the conversation into a debate about Kafka's merits versus RabbitMQ or Pulsar, instead of focusing on the underlying architectural principles. how do you strike the right balance? when is it appropriate to name specific technologies, and when should you stick to more abstract concepts like 'a message queue' or 'a distributed cache'? i want to demonstrate practical knowledge without getting sidetracked.