Summary
Steefan Contractor is a data scientist and AI engineer with 11 years’ experience applying hybrid deep probabilistic models to real-world spatiotemporal problems in climate, oceanography, and remote sensing. Based at UNSW’s School of Mathematics & Statistics and now also working with CounterCurrent AI, he blends deep learning, Bayesian inference, and generative modelling to create robust gap-filling, forecasting, and uncertainty-aware systems for large oceanographic and satellite datasets. His PhD produced REGEN, a globally gridded daily precipitation dataset and non-parametric tools for detecting distributional changes, demonstrating his strength in building reproducible, high-impact scientific datasets and methods. He has delivered end-to-end pipelines and real-time visualization tools (including Shiny apps) and developed LSTM-based temperature gap-fillers with Monte Carlo dropout to quantify uncertainty — a practical edge often missing in applied ML. Based in Sydney, he bridges academic research and industry deployment, focusing on scalable, interpretable solutions for environmental science.
11 years of coding experience
2 years of employment as a software developer
UNSW Sydney