Scale By The Bay 2018: Robert J Neal Interview

yeah I'm Robert Neal I work at udemy as yeah I'm Robert Neal I work at udemy as a principal software engineer on the a principal software engineer on the data team and we're here in San data team and we're here in San Francisco right now our focus is on a Francisco right now our focus is on a Bayesian experimentation platform so Bayesian experimentation platform so it's fairly unique in that not a lot of it's fairly unique in that not a lot of companies are doing Bayesian companies are doing Bayesian experimentation at scales so there's a experimentation at scales so there's a lot of interesting challenges there and lot of interesting challenges there and yeah I mean there's challenges both on yeah I mean there's challenges both on the framework side as well as the the framework side as well as the analysis side and the education side so analysis side and the education side so challenge is all around but it makes for challenge is all around but it makes for really useful business value output yeah so I guess the most interesting yeah so I guess the most interesting track is probably gonna be it's hard to track is probably gonna be it's hard to pick pick maybe software architecture yeah so maybe software architecture yeah so today we'll be talking about how to today we'll be talking about how to efficiently do bootstrapping at scale so efficiently do bootstrapping at scale so bootstrapping is a statistical method bootstrapping is a statistical method it's fairly computationally expensive it's fairly computationally expensive but you know with appropriate algorithms but you know with appropriate algorithms and using spark and Scala I you can do and using spark and Scala I you can do it fairly efficiently so we'll be it fairly efficiently so we'll be talking about that today that's a good talking about that today that's a good question yeah I mean I guess they're you question yeah I mean I guess they're you know everybody's doing know everybody's doing you know very interesting and different you know very interesting and different things so there's like a lot of things so there's like a lot of opportunities to learn opportunities to learn [Music] [Music] poverty's so far there's been shared poverty's so far there's been shared some interesting talks on just some interesting talks on just functional programming generally on functional programming generally on things like acha acha streams and those things like acha acha streams and those are all I think things that we can bring are all I think things that we can bring back to udemy and and use to help our back to udemy and and use to help our business for me functional programming is a lot for me functional programming is a lot easier to reason about so my background easier to reason about so my background is primarily in object-oriented is primarily in object-oriented programming for the first decade of my programming for the first decade of my career and then once I was exposed to career and then once I was exposed to Scala I never looked back I think you Scala I never looked back I think you could just write programs that are much could just write programs that are much easier to reason about express things easier to reason about express things more concisely and focus more on solving more concisely and focus more on solving really interesting problems rather than really interesting problems rather than you know struggling with your your you know struggling with your your programming language or your environment programming language or your environment now a lot of programming languages are now a lot of programming languages are moving more towards conciseness but I moving more towards conciseness but I think functional programming still has think functional programming still has an edge on reasonability an edge on reasonability and is a natural way to express and is a natural way to express solutions to problems I mean yeah everyone at scale by the bay I mean yeah everyone at scale by the bay has been great there's been a lot of has been great there's been a lot of good talks and in talking to you know good talks and in talking to you know people who aren't even giving talks are people who aren't even giving talks are also doing interesting things so I mean also doing interesting things so I mean just like the wide range of people who just like the wide range of people who are working on different things and are working on different things and functional programming I had lunch with functional programming I had lunch with contributor to cats and got a lot of contributor to cats and got a lot of interesting information about where cats interesting information about where cats is going some of the differences between is going some of the differences between scat cats and Scala said so just like scat cats and Scala said so just like really interesting opportunities to really interesting opportunities to learn what people are interested in what learn what people are interested in what they're doing and how you can apply that they're doing and how you can apply that to the work [Music]