An “in the trenches” interview on building a machine learning application with Rails & Hadoop. During the interview on FlightCaster, Brad describes some of the challenges of working with flight data, statistical approaches for flight prediction, false negatives in FlightCaster, Clojure, Hadoop & Amazon EC2, YCombinator, and more. Was pleasantly surprised at how open Brad was about the model internals and data crunching pitfalls.
Comments
An “in the trenches” interview on building a machine learning application with Rails & Hadoop. During the interview on FlightCaster, Brad describes some of the challenges of working with flight data, statistical approaches for flight prediction, false negatives in FlightCaster, Clojure, Hadoop & Amazon EC2, YCombinator, and more. Was pleasantly surprised at how open Brad was about the model internals and data crunching pitfalls.