About

I am a PhD student at the Artificial Intelligence Research Unit at the University of Cape Town (UCT). My research uses physics-based machine learning to model and simulate hydroponic plant growth from sensor data, as part of the Mi-Hy project.

More broadly, I am interested in integrating scientific knowledge into machine learning systems – whether through the loss, the inductive biases, or letting models handle heterogeneous data.

Code lives on my GitHub; I'm also on LinkedIn.

Background

I did my undergraduate majoring in Mathematics, Mathematical Statistics and Computer Science at UCT. After this I worked as a software developer at Yellow Africa and as a data scientist at Isazi Consulting. Before returning to do my masters in mathematical sciences at the University of Oxford. My masters was in many different areas of applied mathematics such as networks, geoscience, Monte Carlo methods and computational medicine. My dissertation was creating a differentiable simulation of electrochemical reactions. This enables us to use more sophisticated, gradient-based, sampling methods to estimate the kinetic parameters of species in solution. This had a very serendipitous link to the "Mi" in "Mi-Hy".

Away from Research

I enjoying playing sports like touch rugby, tennis, golf, football… really anything that is on offer.