Hey all. I recently finished writing a book called Experimentation for Engineers: From A/B
testing to Bayesian optimization
I'll be here to discuss the book in a few weeks, but I can tell you right now that it's about the final stage of production-izing any new ML/AI model or code change you come up with: An experiment which measures its business impact.
You can read the book online -- or get a paper copy -- here:
https://www.manning.com/books/experimentation-for-engineers