Jay Kreps interview
summary
Argues that treating data as a continuously updated stream changes architecture and makes real‑time processing practical, and describes concrete pieces built to prove it: Apache Kafka for durable distributed logs, the Kafka Connect framework for moving data in and out, and Kafka Streams for in‑application stream processing, all shipped as part of Confluent and used in production to replace batch pipelines and lower latency; recommends strong math plus lots of hands‑on pipeline work to move from theory to deployable systems.
AI-written summary · about
related talks
50mApplication development and data in the emerging world ...
03h 00mStream Processing, Going Beyond Pub/Sub with Kafka Streams and...
21mREST APIs vs. Data Streams: Which Is Right for Your Business?
39mData Streaming? I don't even know her
45mApplications in the Emerging World of Stream Processing
06mInterview
55mDemystifying Stream Processing with Apache Kafka
16mLife is But a (Data) Stream: Building Quality Data Pipelines