More output does not mean more taste.
When generation is cheap, the scarce work is comparison: seeing dimensions, tradeoffs, and evidence across options.
Design judgment + human-centered AI
Making AI tools that sharpen design judgment.
I build and study interfaces for an AI-heavy design world: tools that help people notice differences, explain tradeoffs, and build taste over time.
Cognitive Science Ph.D. student at UC San Diego Advised by Prof. Steven P. Dow in ProtoLab.
Design loop Vibes -> variables -> value
Research question How can interfaces make better design decisions visible?
Thesis thread
Taste is a research problem.
Scaffolding for Taste asks a simple question: when AI can make ten plausible options in a minute, how do people learn to notice the difference that matters? I study interfaces that turn generative abundance into clearer comparison, stronger discernment, and better reasons for choosing one direction over another.
Design principle
Design help should not just make more artifacts. It should make the next judgment easier to explain.
Origin note Don Norman's Design Lab talk helped sharpen this into a question about educating designers for an AI-heavy future.I like applying this to urban design, architecture, interior and product design, fashion, robotics, education, and creative media: places where ideas move from vibes -> variables -> value.
Why now
Why this work matters now
AI makes it easy to produce variants; my question is how people learn to choose well.
Everyone is closer to the prototype.
Designers, researchers, PMs, and engineers all shape early artifacts now. Interfaces need to support shared reasoning before decisions harden.
Speed still needs reflection.
Fast cycles are useful only if teams keep room for critique, incubation, and explaining why a direction is better.
Research focus
Design, evaluate, situate.
A small interactive sketch of the loops I keep studying: generate directions, compare what changed, read evidence, then ground assistance in the situation where work actually happens.
Showing the Design motion sketch.
Motion note Inspired by motion craft from Stripe's design team and Katie Dill at Stripe Sessions.
Design lens
Human-AI for design and taste
I study interfaces that scaffold creative judgment: helping people explore and refine generative possibilities without losing the dimensions, constraints, and values that make a design worth pursuing.
See DesignWeaver Moves
- 01
Ground generation in constraints, values, and criteria.
- 02
Make differences visible enough to compare.
- 03
Turn vague preference into evidence-backed direction.
Evaluate lens
Methods for evidence and iteration
I care about what changes after a tool enters the loop: diversity, verification cost, calibration, appropriate reliance, and whether traces help people revise the artifact after deployment.
Read CHI 2026 workshop paper Evidence loop
- 01
Build systems, then measure how people revise with them.
- 02
Treat traces, probes, and metrics as design evidence.
- 03
Use evaluation to decide what should change next.
Situated lens
Embodied systems and mixed reality
This is ongoing work around systems that support everyday tasks through in-situ intelligence, spatial interaction, and real-world use, where the right assistance depends on the person, task, tool, and physical or social context.
Situations
- 01
Let context change the shape of the interface.
- 02
Keep people, tasks, tools, and physical space in the model.
- 03
Make assistance fit the situation instead of interrupting it.
Exploring
Adjacent questions I keep circling
Secondary threads I want to keep prototyping around, while the thesis stays centered on design judgment.
Everyday robots Routines, homes, shared spaces, uncertainty, and unobtrusive help. Learner-centered AI Education at scale while preserving agency and the learning process. Post-deployment iteration Reading real-use traces to improve benefit and reduce harm. Community-centered tools Coordination, sensemaking, and shared ownership in real groups.
Selected publications
Papers worth opening first.
- Short form Workshop First author 0 citesIn Herding CATs: Making Sense of Creative Activity Traces (CHI 2026 Workshop), Feb 2026Workshop Position Paper
Updates
Recent signals.
Pinned Met the great Don Norman and even got a selfie + a “To Sirui” signed “Yellow Book”! 
Our CHI 2026 workshop position paper, What Happened and Why? Trace-Guided Micro-Episodes with Elicited User Explanations for Product Iteration, was accepted to Herding CATs - easily one of the best workshop names ever! The whole ProtoLab is going to Spain! Might be a UCSD party! 
HotSpot got selected as a CVPR 25 Highlight!
Research opportunities
For students who like messy questions.
I like working with curious, motivated, and kind undergraduate and master's students, especially people who have a question they cannot stop poking at.
You do not need to show up as a polished researcher. It helps if you like reading carefully, making small prototypes, testing claims, looking honestly at evidence, and writing clearly about what changed.
The best fit is someone with real stake in a domain or problem, plus enough patience to turn that interest into a concrete study.
Reading closely Prototyping Evaluation Analysis Communication
Start here
- Read the research-start post.
- Email s1tao@ucsd.edu with subject "UCSD Research Interest".
- Include a 1-page CV or resume, an optional portfolio link, and a 3-5 sentence note on the questions or domains you care about.
Connect
Find me around the web.
Advisor Lineage & Past Collaborations
Ph.D. Steven P. Dow HCI
Master's Steven P. Dow HCI Tzu-Mao Li Graphics
Undergrad Judith E. Fan Cognition & Intuitive Physics