Pedagogical Practice

This is a lower-division elective open to students across disciplines. Most arrive without prior coding or electronics experience, and some arrive wary of both. I designed the course around that starting point.

Rather than treating physical prototyping as a technical curriculum to be covered, I use it as a way to study behavior. Students work with sensors, microcontrollers, and simple conditional logic to build small systems that respond to their environment. I frame these systems less as machines than as organisms: things with states, tendencies, and something resembling personality.

The underlying exercise is decomposition. A student begins with a behavior they want to see in the world, such as a creature that grows calmer when approached slowly, or a surface that notices when it is being watched, and works backward into inputs, processes, and outputs, searching for the smallest set of rules that can reproduce it. Programming, in this framing, becomes a medium for describing behavior rather than an end in itself.

What I hope students take away is less a specific toolchain and more a way of thinking: that complex intentions can be broken down, that logic is expressive, and that the gap between "I want it to feel like this" and "here is how it works" is something they can close on their own.

This is an upper-division studio for Interaction Design majors. By this point students are fluent in screen-based interaction patterns, and the course is built to push past them.

I center the class on Unity as an interaction environment rather than a rendering one. Students work with events, states, modular logic, and external inputs (custom hardware over serial, conversational agents wired in as runtime components) to assemble prototypes that extend beyond the screen. The orientation is close in spirit to the Alt.Ctrl tradition: interaction is treated as something composed from inputs and constraints, not selected from a menu of standard UI patterns.

Much of my attention as an instructor goes to the decisions that hold a system together once it grows past a single feature. Clarity of interaction intent, robustness under real conditions, legibility of the underlying structure: these are the skills I believe separate a prototype that demonstrates an idea from one that can actually be tested, critiqued, and iterated on.

Generative and dialog-based AI enter the course as another class of input and output: unreliable, expressive, and worth designing around rather than designing to hide. I want students to leave with the confidence to treat unfamiliar technologies, whatever comes next after LLMs, as material they can learn quickly, reason about, and compose into something of their own.