skip to content
John Sietsma

About

I like to make things, mostly in 3D (games, XR, point clouds, Gaussian splats). Lately that means 3D capture and machine learning. I’ve been a programmer, a technical lead, a co-founder, a teacher and a technical evangelist, and the thread through all of it is building something and then showing people how it works.

I started in AI, back when that meant something different. I helped build a neural-network search engine that answered plain-English questions, did a Master’s in Intelligent Systems, and wrote a thesis on giving game characters emotions.

Then I spent a long time making games. I ran the tech team at Transmission Games, where we rebuilt London from OpenStreetMap data. I led programming at Big Ant. I was a co-director at League of Geeks working on Armello. I was the only programmer on Oscura Lost Light. Games taught me about performance and hardware as well as psychology and human experience.

After that I got interested in putting the real world into the picture. At Dekko in San Francisco I built Augmented Reality on the iPhone before ARKit existed: no markers, no depth sensor, just the camera. At Many Monkeys I wired rooms full of Kinects together to turn spaces into playgrounds. At the University of Newcastle I built VR classrooms, a crime scene for criminology students and a radiation lab for physics.

For a few years I was Unity’s Technical Evangelist for Australia and New Zealand. I wrote shaders that still ship in AR Foundation, helped run a 500-person conference three years running, and taught rendering to everyone from primary school kids to conference halls. I specialised in XR and worked closely with Microsoft and the Hololens.

More recently I’ve worked closer to the data. I led a team processing rail point clouds at scale, then did a run of machine learning projects: tracking players from broadcast footage, measuring athletes from phone video, and an acoustics recogniser that could be fine-tuned by audio experts for different applications. That one became a paper at Inter-Noise 2026.

Now I’m building CultCap: a phone app and pipeline that turns a walk around an object into detailed 3D. I want a regional museum to be able to put its collection online without a $50,000 scanner. I’ve written about why I’m doing this.

I live in Newcastle, Australia. I build in the open and write here about what I learn.

Publications

Clayton Sparke and John Sietsma. An Update on Evaluating Sound Pressure Levels of Noise Sources in Busy Noise Environments with Deep Learning. Proceedings of INTER-NOISE 2026, Adelaide, August 2026.

Patent applications

Elsewhere