Melvin is a Data Scientist within the South African lab of IBM Research. He is primarily working within the Cognitive Astronomy group; whose close collaboration with the flagship SKA project provides the opportunity to contribute towards groundbreaking, fundamental astronomy research. Melvin is developing computationally efficient techniques to automatically detect anomalous astronomical objects within streaming datasets. He has contributed towards the development of statistical classification techniques for supernovae light curves as well as the implementation of Bayesian methods in order to fit cosmological models to supernovae datasets. He also holds an Adjunct position at the University of Cape Town.
Before joining IBM, Melvin worked within academia. He lectured for five years at the School of Statistics and Actuarial Science at the University of the Witwatersrand and subsequently worked for eight years within the Statistical Sciences department at the University of Cape Town. Apart from his research within cosmology, Melvin has developed inferential techniques for stochastic models as well as investigated the dynamics of ecological populations. To date, he has produced fifteen peer-reviewed publications. He is an experienced and successful postgraduate supervisor with three of his students winning five national awards for their postgraduate work. A collaboration with one of his PhD students has produced three R packages on Stochastic Processes.
Melvin holds a PhD in Statistics from the University of Cape Town and a Diploma in Actuarial Techniques from the Institute and Faculty of Actuaries within the UK. He is a two-time winner of the Herbert Sichel medal (awarded by the South African Statistical Association for the best Statistics paper authored by a South African researcher over the past year) and the Claude Leon merit award (awarded by the University of Cape Town for meritorious scholarly work by an early-career researcher).