MLOps.community  cover image

Learning from 150 Successful ML-enabled Products at Booking.com // Pablo Estevez // Coffee Sessions #46

MLOps.community

00:00

How to Support the Plasform Rise?

If your interfaces are simple, you can solve that problem of maintain ability verse generalize ability. If whatever modelling approach yre doing ends up in a table of liner coefficient, then i make a jaba base linear cofice and an engine that only has olap,. And then we us ta going deper iffu of something like pie ortur pencer flow. They have their own run time engines, so you can do whatever you like. Try to support as many se cases while keeping ido the plato. We got through all the lessons, and we got two more. That's eight lessons from booking dot com on how to serve a hundred, 50 machine learning models.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app