Kafka in the Cloud: Why it’s 10x better with Confluent | Find out more
Data systems have mostly focused on the passive storage of data. Phrases like “data warehouse” or “data lake” or even the ubiquitous “data store” all evoke places data goes to […]
Today I’m excited to announce that Neha Narkhede, Jun Rao, and I are forming a start-up called Confluent around realtime data and the open source project Apache Kafka® that we […]