05 papers, published in full

The next frontier of intelligence is visual. Plattipus puts human creativity first, and all of our research exists to amplify it, so that imagination has no ceiling. The research spans AI and machine learning, computer graphics and simulation, computer vision, physical AI and the tools artists work with, and much of it sits where those fields meet. Each method is tested on a working production before it is published. What we publish, we publish in full, and the tools we release are released complete.

Show···
R-01

AI Media Literacy

In October 2026 Plattipus ran a short-form campaign on how to tell real images and video from AI, and it passed 1.4 million views on Instagram in under 20 days. This is the full guide behind it: twelve tells you can check yourself, why each one exists, which ones are already fading, and what the numbers taught us about how people want to learn this.PUB · OCT 2026CAT · AI & storytelling
R-02

Warp at the Geometry Boundary

GPU kernel languages have made the compute half of procedural simulation close to free. We built a compiled Houdini operator that executes NVIDIA Warp kernels directly on geometry, then measured where the time goes. Across every workload we tested, the kernel was never the cost. The boundary between the host application and the compute layer was, by an order of magnitude or more, and it is the part almost nobody instruments.PUB · SEP 2026CAT · Tools & open sourceLIC · OPEN SOURCE
R-03

AI Slop Is a Craft Problem, Not a Tech Problem

We are a production and research house that spent years learning to make frames that hold up, and we build and use AI to amplify that craft, never to replace the judgement behind it. This is our position on AI slop, from the floor of a working production.PUB · AUG 2026CAT · Essays
R-04

Crucible

Every procedural effect in a Houdini pipeline passes through an irreversible step: the bake that turns a live asset into geometry a renderer will accept. We argue that step is in the wrong place. Crucible is an open-source Hydra scene index plugin that moves it to render time, cooking HDAs lazily as Karma pulls on the scene graph, with no pre-cook, no LOPs recook and no HAPI session to manage.PUB · MAY 2026CAT · Tools & open sourceLIC · OPEN SOURCE
R-05

A Splat Should Be a Primitive

Gaussian splatting produces photographic reconstructions faster than modelling can, and then strands them outside the pipeline as a foreign format that composes with nothing. We argue the fix is not a better importer but a change of status: define splats as first-class USD primitives with their own schema, imaging adapter and Hydra delegate, and every composition feature in USD applies to them for free.PUB · MAY 2026CAT · Tools & open sourceLIC · OPEN SOURCE