Talking about how Klarna is organized to support efficient data science and how the backend design has played a big role both for automation and to create clear division of responsibilities between data engineers and data scientists.
- How do you create an environment with short time to market for machine learning models? Why is it important?
- How differences in analytics and live environments is a great challenge in data science?
- Having one code base for features that is the same for analytics and live is a big step forward. I will show how it can be enabled by creating a common event based data model
- How Data Scientists are empowered by owning feature code base?