My primary experience is in applied deep learning and computer vision. I have worked across the full stack of machine learning stages, including working with raw data and setting up (multi-node) infrastructure for training / validation / deployment. My programming language of choice is Python (although occasionally I also write in C++ and CUDA if needed) and my deep learning framework of choice is PyTorch (although I have worked with TensorFlow and some of my past work is still quite popular). For deployment purposes, I have primarily worked with TensorFlow.js and TensorRT.
I am currently a senior data scientist of the AI team at Nearmap.
Before that, I was leading deep learning efforts and managing a small team at Skydio, where I mainly concentrated on building end-to-end reproducible deep learning pipelines.
I did my PhD studies at the University of Adelaide. I was fortunate to work there under the supervision of Prof. Ian Reid and co-supervision of Prof. Chunhua Shen. During my doctorate studies I used deep learning to solve various dense per-pixel computer vision tasks (primarily, semantic segmentation) in real-time and with high performance. My doctorate thesis is available through the University of Adelaide library service. As a PhD student, I was the first author of 6 academic papers, 5 of which were published and presented at major computer vision and robotic conferences, including CVPR and ICRA.
I believe in hard work and curious mind.
My major hobby is European football (known to some by an s-word) and my favourite team is Liverpool FC. I have been supporting Liverpool since 2007 and so far have been fortunate to see them twice with my own eyes (both times losing, so I might be an unlucky charm ).
My other hobbies include reading books, watching movies and listening to music.
My CV is hosted here