Change management in data teams can be challenging to say the least. Not only you have to evolve your data pipelines, data structures, and datasets themselves across environments, you also have to keep data exploration and visualizations tools in sync.
In this talk, we’ll be exploring how to do this best across environments (ie: dev, staging and prod), talking about how CI/CD can help, implementing good data ops practices and cranking up the level of rigor where it matters.
We’ll also talk about rigor-vs-speed tradeoffs, where clearly not all data pipelines are born equal, and how to think about to evolve the level of rigor over time in places where it matters most.
