Scale By The Bay 2018: Julie Pitt Interview

Recording: Scale By The Bay 2018: Julie Pitt Interview

I'm Julie Pitt and I'm at Netflix leading the machine learning infrastructure team and I'm from the East Bay Livermore California building a machine learning platform so that data scientists can focus on data science and not all of that messy engineering work underneath so the coolest thing about it is that they can get access to large amounts of compute resources early in the development cycle and they don't need to even think about scaling their machine learning models data pipelines hands down very relevant to anybody doing machine learning they all need data thinking about how to build software in a use case driven manner really diving into in our case building a platform and how our data scientists actually getting their job done and how do we use that to inform what we add to the platform I'd like to learn about best practices around new and emerging technology in machine learning so one thing that's interesting with the auto ml or how to do hyper parameter optimization scala made me a lazy developer because the compiler took care of a lot of things that I used to have to deal with I think over the next 15 years Scala has a challenge and I think that the Scala team is actually addressing it which is how to keep it very simple and light while it's getting increasing adoption has become mainstream though there's so many of them I love coming back to this conference every year and there's a good core group of folks that come back so just said hi to dick wall and iacopo turski and another friend of mine Ely it's just so great to come together not any one person but it's really the community around it