Pixel Surgeon at the intersection of art, code, and machine learning.
Mike is an artist–engineer who creates unseen images using computational means. He works closely with teams on cinematic and technically demanding visuals, believing that ambitious results come from people who genuinely care about the craft and are willing to push through hard problems together.
For personal work, Mike starts with a visual idea and then builds dedicated software specifically for that idea. The process is iterative — the image and the tool develop in parallel. At every step, the artistic goal and the technical constraints push against each other. Mike treats that friction as useful information — a signal about what needs to be redesigned or rethought. The result is images that could not have been made with off-the-shelf tools, because the tools did not exist until the work demanded them. He prefers to present finished work as prints or other low-energy formats whenever possible.
In a VFX team, Mike tends to take on the toolsmith role. Because he approaches problems as an artist first, the tools he builds reflect that — parameters that map to how artists think, interfaces that stay out of the way, and workflows that let the team focus on the image rather than the software. Strong tooling is not a side task; it is what makes meaningful visual quality achievable within a fixed schedule.
Mike holds a Ph.D. in Computer Science and Engineering, with a focus on deep learning for computer graphics. After a long period based in Hong Kong, he is now actively looking for overseas roles where he can contribute in a hands-on capacity — specifically in ML-driven digital image synthesis and its application in real production environments. His focus is on making the gap between cutting-edge methods and what actually ships as small as possible.
I am on X , LinkedIn and Github.
Most of my research pre-prints are available on my ResearchGate page.