Just when you thought that you had enough things to learn to keep from getting too far behind in the new technology stacks for deploying applications, comes the PANCAKE STACK (see pancake-stack.com). The PANCAKE-STACK spans a wide range of data science and engineering components for one to evaluate or adopt in a Spark-based end-to-end recommendation engine and data pipeline.
Chris Fregly, formerly the Lead Streaming Data Engineer at Netflix and now at PipelineIO came up with the acronym. He received an Emmy for Streaming Engineering Excellence while at Netflix. The idea for PipelineIO dates back to his time at Netflix where they were forced to build a custom machine learning prediction/serving layer. There wasn’t (and still isn’t, in his opinion) a production-ready, fault-tolerant, and low latency open source system to serve Netflix-scale predictions and recommendations in real-time.