About the event
We’re excited to have our first in-person event in Auckland, New Zealand! The Modern Data Stack Chat is hosted by our friends at Countdown. This in-person event is meant to bring together the local data & analytics community in Auckland. Have a few slices of pizza, drink and network with other data enthusiasts and find out more about the Modern Data Stack. We’ll also have great speakers throughout the evening.
When and where
- Location: CDX office, Level 8, 4 Williamson Avenue, Grey Lynn, Auckland 1021
- Time: 5.30PM NZDT - 7.30PM NZDT (first speaker at 6.00PM)
- Date: 16 February, 2023
Anna Choi (Data Engineer - Analytics & Insights Team at Countdown)
- Talk Title: (Re)building feature store with dbt
- Talk Summary: Are your stakeholders tired of getting different answers for the same analytics question?
This talk will walk you through the journey that the Analytics & Insights team at Countdown’s had to build a robust, re-usable feature store - from the first attempt of python spaghetti horror to the current state using dbt for better maintenance and visibility.
Diego Morales (Analytics and Insights Practice Lead at Watercare)
- Talk Title: What does embracing a modern data stack looks like in the Water Industry?
- Talk Summary: In this presentation I am aiming to take you through the journey of how Watercare has designed and implemented his analytics data stack, what challenges we are still facing and working towards solving, and how dbt has helped us to transition from almost impossible to maintain code to a fully productionized, scalable data warehouse designed for self-service, with a completely automated deployment pipeline.
Joel Labes (Senior Developer Experience Advocate at dbt Labs)
- Talk Title: Using Slim CI to speed up your iteration cycle during development
- Talk Summary: Before joining dbt Labs, Joel was Head of Data at NZ-based edtech tool Education Perfect. He will share his experience using Slim CI to speed up EP’s continuous integration process by 10x, and how combining Slim CI with exposures prevented changes to models from unexpectedly breaking dashboards in production.