Infra at Netflix

a common ML infrastructure which would help our data scientists to iterate on these projects faster and more effectively.

Hypothesis:

Our working hypothesis is that there isn’t a one-size-fits-all approach when it comes to ML testing or monitoring. Instead, we want to provide easy building blocks and patterns for data scientists which they can use to tailor a testing approach that works for their use case.

How to build an infra?

Summarize the main issues. 1. Optimize elements 2. Enhance quality of streaming

• Could have multiple infra for different use case (high volume / wide variety etc)

• MetaFlow