#sivart — the designer

applied AI code & design systems typography since ~2004 6 active R&D

travis reversed. Two decades of practice across applied AI, code, generative systems, typography, and packaging — in the Inland Northwest. Where the building lives.

artist educating design.

##practice areas

id	area	since · status
S1	applied artificial intelligence	2018 · active
S2	emergent & generative design	2020 · active
S3	design systems & foundry	2024 · active
S4	agent tooling & marketplaces	2025 · active
S5	packaging design	2008 · periodic

##recent work

2024 — 2025. Selected projects.

year	project
2025	Cybersecurity Department wall graphic. Designed and coordinated production of a 23′ × 8′ installation at EWU. Installed Aug 2025.
2025	Itron donor recognition sign. Faculty lead with EWU Foundation; directed student design team through concept, prototyping, and fabrication of a 7′ × 5′ sign.
2024–25	Spokane Scholars Foundation rebrand. Faculty lead on website enhancement and rebranding roadmap; UI work, donor strategy, student-coursework integration.
In dev	"Who Does It Think We Are?" Interdisciplinary research with Charlie Potter, Justin Young, and Marielle Leijten on user persona & story development in GenAI contexts.

##r&d — overview

R&D is part of an ongoing investigation into how agents and humans can do research-to-production work together: where AI takes the wheel, where it stays in the passenger seat, and where the boundary moves over time. The six projects below each press on a different facet of that question.

N° 01

agent adworks

active

a marketplace experiment · 402c protocol

Agents bid on tasks. The platform ranks by quality and price. Reputation becomes market infrastructure. The interesting question is how to evaluate agent quality, bid truthfulness, escrow, and reputation when output quality is only known after delivery.

Core ask: How should agent quality, bid truthfulness, escrow, and reputation be evaluated?
Role of AI: Architecture, implementation, docs, SDK design, protocol critique.

N° 02

signal desk

paused

a personal briefing system

A source-first AI information hub that turns feeds into a briefing: what matters, why it matters, what the catch is, whether to click. Currently paused while I work out an ingestion architecture that won't blow the budget.

Core ask: How do I re-enable AI ingestion with batching, call caps, prompt caching, and safe backfills?
Role of AI: Classification, synopsis generation, ranking logic, reader-personalized briefings.

N° 03

program command

active

academic schedule planning

A planning system for enrollment trends, faculty workload, capacity, department profiles, and schedule-builder decisions. The premise: agent workflows can support high-stakes academic planning, but only if data assumptions stay visible the whole way through.

Core ask: How can agent workflows support high-stakes planning without hiding data assumptions?
Role of AI: Data QA, planning docs, interface iteration, scheduling logic, verification scripts.

N° 04

ai + design / canvas

active

course system · DESN 374

Canvas-ready DESN 374 materials, AI inquiry studio assignments, student evaluation reports, and a career kit for translating AI work into portfolio language. The question is how to teach agentic workflows without turning the class into tool-chasing.

Core ask: How do I teach agentic workflows without turning the class into tool-chasing?
Role of AI: Curriculum design, critique rubrics, feedback drafts, project briefs, student-facing language.

N° 05

canvas pedagogy analysis

active

a corpus over years of teaching

A pipeline over years of Canvas exports: corpus indexing, content extraction, coding schemas, QA reports, theory memos, and privacy guardrails. The agent has to preserve evidence, citations, and instructor voice — not flatten them into average prose.

Core ask: How should a research agent preserve evidence, citations, privacy, and instructor voice?
Role of AI: Corpus navigation, synthesis, methodological critique, evidence-backed memo writing.

N° 06

uigen

prototype

a component generator prototype

A local AI-powered React component generator with live preview, file persistence, and Claude-backed iteration when an API key is available. Less a product than a teaching artifact: what makes AI-assisted interface generation inspectable, editable, and pedagogically useful?

Core ask: What makes AI-assisted interface generation inspectable, editable, and pedagogically useful?
Role of AI: Component generation, code explanation, iteration, exportable implementation.

##open questions, highest leverage

What I'm chasing.

#	question
01	Research agents. How to structure a long-running research agent over private Canvas archives with evidence, citations, privacy, and interpretive voice intact.
02	Ingestion at low cost. The right architecture for batching, prompt caching, call caps, retries, dedupe, and local-vs-API enrichment in a personal briefing system.
03	Agent marketplaces. Evaluation and reputation patterns when agents bid on tasks and quality is only known after delivery.
04	Teaching workflows. Patterns for human-in-the-loop agents in education — assessment, critique, and curriculum design without tool-chasing.
05	Project memory. How to separate Projects, Code repos, skills, MCP tools, and documentation so work doesn't disappear into chat history.

— end of file —