I’m a machine learning manager, researcher and software engineer with 14+ years of experience in developing and improving state-of-the art machine-learning research algorithms and turning them into useful, reusable code that can make life easier for researchers, software engineers, and everyone else!
I’m now working with TAFE NSW, leading the Learning Analytics Team in their work to take a data-science approach, including AI and ML techniques, to improve the learning experiences for vocational training across the entirety of TAFE NSW.
When I was still in Brisbane, I ran local events for the OpenStreetMap community. So, if you came here looking for OpenStreetMap events in Brisbane, please go have a look at the Brisbane Events Page on the OSM Wiki, or my EventBrite page.
Armidale is a bit quiet for this sort of thing, but it would be great to get similar events running here.
OpenStreetMap and Smart-GIS: I’ve been an OpenStreetMap member since 2007, and was responsible for a lot of the original mapping and social events in Brisbane, Australia. I want to get more into AI and machine-learning for GIS applications to help the OpenStreetMap community.
I have enjoyed supervising 7 PhD students over the past 7 years, and have done occasional seasonal academic work over that time as well.
Throughout 2017, I worked with the University of Queensland developing an exciting new Masters of Data Science, and I enjoyed developing datasets and teaching material that helps in the training of future professional data scientists.
I have also done some mentoring with CoderDojo Brisbane, and helping kids get excited about coding is pretty awesome.
In the past, I have done some machine-learning consultation work with two exciting medical machine-learning start-ups, Wink Health in California, and M3dicine in Brisbane. It’s great to be able to help medical professionals and the general population to get access to improved health care through innovative signal processing and machine learning.
I have been working with audio-visual speech since 2004, and loved showing that visual information is almost always complementary to acoustic over a wide range of applications.
Speaker recognition and diarisation: A large part of my post-doctoral experience since completing my PhD in 2008 has focused on conducting research, development and commercialisation of state of the art speaker recognition and diarisation research.
As of November 2017, I have 771 citations across more than 80 publications, with 24 publications having more than 10 citations, and a h-index of 14. More details, and a full list of publications can be found on my Google Scholar profile.
Publication venues include: