Rhimmon Simchy-Gross, PhD UX Research Portfolio

Rhimmon Simchy-Gross, PhD

I build the human-AI research systems that make synthesis fast and trustworthy, the interactive research prototypes and explorable artifacts that put insights in people’s hands, and the executive narratives that connect UX evidence to product decisions and business outcomes.

How this number is counted

{{ statDetail }}

Selected work
Preview of the Misclick roadmap model Preview of the Misclick roadmap model

Measured evidence. Modeled direction.

Interactive quantitative roadmap model with usability case study impact.

Representative mockup · live ↗
Preview of the concept test readout Preview of the concept test readout

Concept tested. Direction set.

A moderated concept test, delivered as the readout that set the experience direction.

Representative mockup · live ↗
Preview of Executive storytelling Preview of Executive storytelling

Executive storytelling.

Eight-beat narrative built around participant and business videos.

Custom interactive experience ↗
Top of the AI-assisted research case

AI-assisted research, human-owned.

Interactive walkthrough of how I work with AI: framing, verification, and human judgment.

Custom interactive experience ↗
Preview of Four deliverables, one interactive artifact Preview of Four deliverables, one interactive artifact

Four deliverables. One interactive artifact.

A directional pulse study on home cooks and AI: readout deck, one-pager, full report, and evidence in one experience stakeholders actually explore.

Representative mockup · live ↗
Preview of the signup triage workflow Preview of the signup triage workflow

AI surfaces. Humans review.

Suspicious signups surface themselves. Reviewers see only the calls that need judgment.

Representative mockup · live ↗
Preview of the Streakwise habit app Preview of the Streakwise habit app

0→1 in 2 hours. Transferable capability.

A functioning prototype, built end to end in 2 hours to show researchers what AI-assisted development puts within reach.

Representative mockup · live ↗
About

My foundation is experimental psychology: distinguishing observable behaviors from the underlying constructs they may reflect, then testing competing explanations.

My PhD on how people perceive, judge, and make decisions led to years of mixed-methods studies and UX product research.

I bring that rigor to AI-assisted research and product work, accelerating throughput while keeping verification, judgment, and accountability human‑owned.

I size studies to the decision and match claims to what the sample supports, stating the unit of analysis, the repeated-measures structure, and the uncertainty.

← All work
‹PreviousAI surfaces. Humans review.
›NextFour deliverables. One interactive artifact.

AI-assisted research, human-owned.

My personal operating model for human-AI research: how I frame the work, check it against reality, and keep the final call with a person.

Research synthesis now runs in ~1/5 the time, every claim traceable to its source and verified by the researcher before it affects product decisions.

A Design Director sent her researcher to learn my method, a Lead Researcher co-built on it, and Research Ops solicited my input on AI tooling.

The operating loop

{{ loopTitle }}

{{ loopCopy }}

Review gates

Risk should change the path.

{{ gateStatus }}

{{ gateTitle }}

{{ gateCopy }}

Two scales

Personal operating model

Frame the problem, route to the right model, specify what success requires, execute, verify, judge independently, refine, and only then accept.

Principles that travel

The same discipline works for any team: sources stay traceable, scope stays visible, and anything consequential gets a human decision.

{{ trustDetailLabel }}

{{ trustDetail }}

Why this scales

Quality review gates encourage other researchers to learn the method and trust the process.

‹PreviousAI surfaces. Humans review.
›NextFour deliverables. One interactive artifact.
← All work
‹Previous{{ artifactPrevName }}
›Next{{ artifactNextName }}

{{ artifactTitle }}

{{ artifactThesis }}

{{ artifactImpact }}

{{ artifactImpact2 }}

{{ artifactImpact3 }}

Also on this page A usability study comparing two payment paths
{{ artifactLabel }} {{ artifactOpenLabel }}
Illustrative data 20 participants · 120 task attempts repeated-measures design 16 to 20 complete pairs across the three direct comparisons
{{ artifactTitle }} Static preview

The 120 attempts come from those 20 people, so they are not 120 independent observations. Direct comparisons are paired within participants, tested with exploratory Wilcoxon and Holm correction across the three. The pooled projection is a negative-binomial GEE with participant clustering. External usability benchmarks are contextual references, not matched control groups.

{{ artifactNote }}

{{ artifactSoWhatLabel }}

{{ artifactSoWhat }}

Supporting quantitative experience

Two payment paths, compared

An unmoderated study design comparing two prototypes on task success,
time on task, perceived ease, overall usability, and open-ended feedback.

Fictional example: 16 of 24 found the payment screen inside the product, yet struggled;
12 of those 16 reached the external payment provider.

Observed path friction select a version
{{ taxLabel }} {{ taxValue }} observed failures reaching the payment provider

{{ taxDetail }}

Preference signal 10 / 12 10 of the successful 12 preferred the cost-preview version
(randomized A/B forced-choice)
Convergence 3 / 3 measures agreed: behavior, preference, and open-ended comments

Illustrative numbers. A fictional example showing how I pair behavioral, preference, and open-ended measures; all figures are invented.

Why this decides

Behaviors, preferences, and open-ended comments all pointed toward clarity and actionability. Convergence across the three measures is what carried the decision.

‹Previous{{ artifactPrevName }}
›Next{{ artifactNextName }}
← All work
‹PreviousFour deliverables. One interactive artifact.
›NextMeasured evidence. Modeled direction.

Executive storytelling.

Reconstruction of a quarterly leadership update: eight beats alternating spoken synthesis with users’ voices. The evidence shown here is illustrative, from a fictional product.

Delivered at a quarterly UX town hall; other teams asked in the chat
for it to become the standard format for every quarterly.

The VP of UX then asked me to present it to Sales, Talent,
and senior engineering and product leaders.

8 narrative beats
8 video clips
3 audiences it serves
1 through-line
{{ beat.label }}
{{ beat.slideAlt }} {{ beat.slideAlt }} {{ beat.slideAlt }} {{ beat.slideAlt }}
Representative slide mockup

{{ beat.title }}

{{ beat.copy }}

▶ {{ beat.mediaLabel }}
{{ beat.clip }}

{{ beat.quote }}

{{ beat.note }}

Why this shape

The alternation is the method.

Spoken synthesis carries the reasoning. Participant audio carries the evidence. Alternating them means a room never has to choose between the two.

Two asks followed: one to brief leadership, one to make the shape the default.

‹PreviousFour deliverables. One interactive artifact.
›NextMeasured evidence. Modeled direction.
Rhimmon Simchy-Gross, PhD Bixby, OK | +1 (315) 481-1422 | rhimmon@me.com rhimmon.com | linkedin.com/in/rhimmon-simchy-gross SUMMARY 13+ years behavioral research; 5 years product research. Tied to multi-million revenue, 1M+ clients, 142 countries. Build responsible, scalable human-AI research systems: ~5x faster synthesis, 2.3x higher output, traceable claims. EXPERIENCE UX Researcher, ADP Remote (Bixby, OK) | 08/2022 - Present • Architect human-AI research practices across our 1K+ studies: source traceability, verification, human review gates, and researcher-owned recommendations, keeping judgment and accountability with the researcher. Synthesis takes ~1/5 the time, every claim traces to source, and throughput rose from 3 to 7 initiatives per quarter. • Lead mixed-methods research end-to-end (2026: 28 interviews, 4 research reviews, 3 surveys, 3 workshops, 2 concept tests, 2 competitive analyses), shaping strategy by reframing enterprise adoption around existing workflows. • Translate research into executive decisions, earning VP of UX recognition for storytelling, plus 4 consecutive requests to present my narratives to Sales, Talent, a VP of Software Engineering, and an SVP of Product Development. • Advance AI product direction across 7 rounds in 2 programs in 2026, shifting an AI feature beyond task completion toward predictable, recoverable behavior; recommendations adopted into the product direction. • Build 8 AI-enabled prototypes, tools, and interactive experiences in 2026, spanning workflow automation, decision-support modeling, explorable research deliverables, and 0-to-1 experience prototyping. • Pioneer the team's responsible AI-for-research practice: authored the analysis framework & source-verification standard with human-review guardrails; taught it through hands-on labs; AI prompts were adapted into official research templates and showcased org-wide, reinforcing human judgment, traceability, and decision ownership. • Scale AI practice across teams: Research Ops recruited me to pilot an agentic research tool and solicited my AI infrastructure recommendations; a Research Director partnered with me to build a process diagram presented to senior leaders; a Design Director sent her researcher to learn my method; and a Lead Researcher brought me in to co-build. User Researcher, TCM Creative: Moore Neuro-Fundraising Lab Tulsa, OK | 08/2021 - 05/2022 • Drove end-to-end primary research across 11 mixed-methods studies, 4 surveys, and 100+ interviews, guiding product decisions across 21 campaigns for 8 nonprofit clients; ad revisions cut cost per call 55%. Postdoctoral Researcher and Assistant Professor, Missouri Western State University St. Joseph, MO | 08/2019 - 07/2021 • Contributed to large-scale quantitative synthesis by aggregating 100+ datasets with 30 scientists across 7 countries. Graduate Researcher and Instructor, University of Arkansas Fayetteville, AR | 08/2013 - 08/2019 • Conducted 40+ NSF-funded studies, 9 articles, 10 global talks. Research featured in Netflix's Explained: Music. • Taught Introduction to Statistics for Psychologists as instructor of record, a required course in inferential statistics. SKILLS Methods: Mixed-methods, qualitative and quantitative research, interviews, survey design, workshops, experimental design Research: Generative & evaluative research, concept & usability tests, competitive analyses, recruitment & screening Statistics: Regression, ANOVA, t-tests, multi-level linear mixed modeling, response effects, data visualizations, R, SPSS Research Tools: Figma, Miro, Mural, Dovetail, Maze, Qualtrics, SurveyMonkey, User Interviews, UserTesting, Pendo Agentic AI: Claude Code, Codex, GitHub Copilot, Microsoft Copilot, VS Code, Python (AI-assisted), agent orchestration EDUCATION PhD, Experimental Psychology | University of Arkansas, Fayetteville, AR | 05/2019 MA, Experimental Psychology | University of Arkansas, Fayetteville, AR | 05/2016 BM, Psychology and Piano Performance | State University of New York at Potsdam, Potsdam, NY | 05/2013