In this project you will explore a series of machine learning methodological developments to facilitate new approaches to insurance product design and pricing. It will be focused on methodological developments that can be applied to a variety of insurance domains. The work will be primarily exploring aspects of kernel machines, boosting and ensemble methods, random forests, Gaussian processes and warped Gaussian processes to enhance data analytics in insurance applications. The considered application domains will be based on real world problems from both general and life insurance applications.