Remove blogs is-the-centralized-data-warehouse-dead
article thumbnail

The fancy data stack—batch version

Christophe Blefari

Summer Edition ( credits ) This is the first article of the Data News Summer Edition: how to build a data platform. The modern data stack has been criticised a lot, a few are saying it's dead other are saying we are in the post-modern era. In this article we will try to design the fancy data stack for a batch usage.

article thumbnail

Internal services pipeline in Analytics Platform

Picnic Engineering

Quick re-cap: the purpose of the internal pipeline is to deliver data from dozens of Picnic back-end services such as warehousing, machine learning models, customers and order status updates. The data is loaded into Snowflake, Picnic’s single source of truth Data Warehouse (DWH).

Kafka 52