Getting Machine Learning applications into production using MLOps and CD4ML – Eric Nagler – XConf NA

According to VentureBeat 87% of data science projects never make it into production. Bringing Machine Learning (ML) models into production is different and more complex than deploying traditional software applications because with ML applications you need to orchestrate and deliver three pieces of software concurrently: the data engineering transformations, the trained ML/AI model, and the software API's that leverage that model. This talk will overview the Thoughtworks' approach to MLOps called CD4ML, and how it can help with ensuring quality, repeatability, and reliability when undertaking Data Science projects.