Specialties & Experience
General Specialties:
Information Technology
Specialty Focus:
Education:
Master of Data Science Bachelor of Engineering (Information Technology)
Years in Practice:
7
Additional Information
Building ML systems at commercial scale and teaching them at university are not the same job. I do both, simultaneously. The gap between how AI is learned and how it actually runs in production is bigger than most people on either side admit.
I lead data science at Sportsbet, where my team ships recommendation and personalisation systems. I also lecture at Torrens University, teaching ML and AI to postgraduate students. Most people at this level move one way or the other. I've kept both roles running in parallel, and the friction between them shapes how I think.
One system I built sits among the first of its kind globally: a deep learning model that detects at-risk user behaviour in real time and automatically intervenes. I deployed it at commercial scale, and this work now shapes how the broader organisation approaches responsible AI.
Deploying an AI model is the easier problem. Making it auditable, defensible, and future-proof after it ships is where most organisations struggle. I've built the production governance layer at commercial scale: monitoring pipelines, decision logging, and the process architecture that keeps AI systems trustworthy over time. Most teams focus on shipping the model. I focus on what happens to it six months later.
At Torrens I teach Responsible AI and its Use in Business, a postgraduate unit covering Responsible AI in practice built to close the gap between textbook ML and production systems.
I lead data science at Sportsbet, where my team ships recommendation and personalisation systems. I also lecture at Torrens University, teaching ML and AI to postgraduate students. Most people at this level move one way or the other. I've kept both roles running in parallel, and the friction between them shapes how I think.
One system I built sits among the first of its kind globally: a deep learning model that detects at-risk user behaviour in real time and automatically intervenes. I deployed it at commercial scale, and this work now shapes how the broader organisation approaches responsible AI.
Deploying an AI model is the easier problem. Making it auditable, defensible, and future-proof after it ships is where most organisations struggle. I've built the production governance layer at commercial scale: monitoring pipelines, decision logging, and the process architecture that keeps AI systems trustworthy over time. Most teams focus on shipping the model. I focus on what happens to it six months later.
At Torrens I teach Responsible AI and its Use in Business, a postgraduate unit covering Responsible AI in practice built to close the gap between textbook ML and production systems.