


AI agents built a 3D printing head animation and scene via text prompts in Houdini, driven by Python-synthesized audio.
Another experiment with Claude and Houdini via MCP.
Everything here was built from text prompts, without touching a single node: modeling, layout, animation, lookdev and rendering.
The sound was synthesized in Python. A second agent then used that audio to drive the animation of the printing head.
Still exploring where this workflow works well and where it doesn’t.
scraped.media
derived.classifier
derived.classifier
scraped.caption
derived.classifier
derived.classifier
user.note
$0.0183
classify $0.0066on-screen-text $0.0117