## Refs [We're still in the steam-powered days of machine learning - Normcore Tech (substack.com)](https://vicki.substack.com/p/were-still-in-the-steam-powered-days) >There really is no good, generalized single system of best practices for creating machine learning platforms. There’s not even a textbook on it. (Other than[ Designing Data-Intensive Applications](https://veekaybee.github.io/2019/04/11/attic-compsci/), which is good but doesn’t necessarily fit the bill.) [[Data architectures]] [Lessons Learned Scaling Machine Learning at Go-Jek on Google Cloud (Cloud Next '18) - YouTube](https://www.youtube.com/watch?v=57Q07-SIcm4) >Give incentives to people to publish data".  >Feature should be free and make them discoverable. > Precompute everything. c