✍️ Building ML Teams and Platforms with Assaf Pinhasi
The MLOps Podcast
00:00
How Did It All Work?
I feel like there wasn't a lot of written content about machine learning in production. So I'm curious, like, how do you build all of this infrastructure and tooling around something where you have no reference point? Sure. It was a very long and challenging process. PayPal grew kind of over time in terms of the number of models, the complexity of the model. Back in 2010, it was a lot of regression, like classical regression models. And, you know, neural networks started coming up maybe in 2013 or 2014. That created a new challenge in terms of compute and in terms of hunger for features to feed into the model.
Transcript
Play full episode
