This was another opportunity to present my work on synthetic insurance data to a broader data science audience. It was also nice to step a little outside the usual actuarial crowd and discuss how statistical and machine learning methods can be used to generate realistic synthetic insurance datasets.