Wray enjoys playing with web/text data, writing code, and devising Bayesian algorithms. He used to support some of his work on Github (wbuntine). His research interests include theoretical and applied work in document and text analysis, machine learning, and probabilistic methods including:  discrete non-parametric and Bayesian statistics with applications to machine learning and deep learning, latent variables in semi-structured and text analysis, and applications in medical and health informatics.  His latest work is using these techniques to develop methods for improved personalised information access.