What needs to be done for Amundsen to be successful at Earnest ?
For example right now we create Glue tables in Haskell, which makes it very hard for analysts to make any type/comment modification on tables, that might create a bad dynamic where analysts have to ping engineers when they need minor metadata changes
We need to think about the obvious (Amundsen on Kubernetes) but also processes that touch metadata in general, also how can we make sure we have a high adoption rate?
What needs to be done for Amundsen to be successful at Earnest ?
For example right now we create Glue tables in Haskell, which makes it very hard for analysts to make any type/comment modification on tables, that might create a bad dynamic where analysts have to ping engineers when they need minor metadata changes
We need to think about the obvious (Amundsen on Kubernetes) but also processes that touch metadata in general, also how can we make sure we have a high adoption rate?