About
South African-born actuary and data scientist with a 10-year track record of driving business value through innovative solutions across multiple industries. Combining a practical problem-solving approach with a strong foundation in applied research, I excel in tackling complex business-driven challenges in collaborative, multi-disciplinary environments.
Experience
As a Head of Product AI at Auto & General, I’m leading a team that is delivering AI solutions for product, underwriting, and pricing, helping to provide affortable and dependable insurance to all Australians.
As a Director of Data Science at Fidelity Investments, I led a talented team focused on personalisation, recommender systems, and marketing optimisation in both digital and distribution channels.
At Quantium, I honed my analytical skills and became proficient in data analysis, statistical modeling, and technical communication, while working on a diverse range of projects across the Insurance, Banking, and Retail industries. Surrounded by a dynamic and talented team at Quantium, I spent five years there building a solid foundation in analytics and data science.
While working, I completed the board exams to become a qualified actuary and receive Fellowship designation from both the Actuarial Society of South Africa and the Actuaries Institute of Australia.
Education
I completed a masters degree in Computational Science and Engineering (CSE) at Harvard School of Engineering and Applied Sciences. During my time at Harvard, I delved into advanced courses, exploring topics such as decision theory, optimisation, and computing systems. My research project involved numerical methods, optimisation, and large-scale parallel computing to study optimal heat transfer solutions in Rayleigh-Bénard convection.
I completed a BSc degree in Actuarial Science and Financial Mathematics at University of Pretoria. The program not only provided me with a rigorous academic foundation in mathematics and statistics, but also cultivated skills needed to develop realistic solutions to complex problems and with a forward-looking outlook on issues.
Projects
White-box feature selection library that supports supervised and unsupervised selection methods. It also provides optimised item selection for exploration in recommender systems based on diversity of text embeddings (via TextWiser) and label coverage by solving a multi-objective optimisation problem (CPAIOR’21, DSO@IJCAI’22).
Research library (IJAIT 2021, ICTAI 2019) written in Python for rapid prototyping of multi-armed bandit algorithms. It supports context-free, parametric and non-parametric contextual bandit models and provides built-in parallelization for both training and testing components.
Recommender Systems library that integrates several other open-source software for building bandit-based algorithms that I presented at All Things Open 2022, Open Data Science 2023 and, NVIDIA RecSys @ Work 2023. See blogpost as a starting point for data scientists to build and deploy bandit-based recommenders using Mab2Rec.
Publications
Interests
Outside of work, I enjoy spending time with my wife and 1-year old boy. We love exploring places, finding good pastries, ice-cream, and friendly dogs to play with. I’m also an avid rugby fan and rock climber (both of which I hope to one day share with my son).