03 / The story
From the classroom
to the stage.
Thunk Machine grew out of Audio Pixel Collider at The University of Texas at Austin and its AET350C Advanced Creative Coding class. Its name nods to a programming idea: a thunk wraps a computation in a function so it can be evaluated later.
Thunk Machine uses computer science as a material for creative work. First-class functions let behavior be passed around and combined; closures retain values; composition and nesting build complex visuals from simpler parts; and objects bring state and behavior together. These concepts become tools for shaping visuals and improvising with code.
These ideas draw on the teaching tradition of Daniel P. Friedman and The Little Schemer, the functional programming lineage of Scheme, and the creative coding foundations of Processing and p5.js. They also draw on work emerging from the algorave and creative coding communities, including Hydra and P5LIVE, and on Char Stiles’s Fantasy IDEs research at the MIT Media Lab, which inspired the integration of conversational AI into the Thunk Machine IDE.
Thunk Machine brings those influences into a performance environment where the underlying program stays visible. It is a place to study an idea, develop a visual system, and improvise with it in front of an audience.




