# Jordan Campbell — public portfolio brief

## Current role

- Title: Director of Product Design & AI Operations
- Company: Inbox Health
- Scope: Sole product designer across product, systems, and internal tools.

I design healthcare payment and AI products, and I ship them in code. Sole product designer across a seven-app platform serving 2.5M+ patients. I start with research, then build in TypeScript and React with Claude Code.

## Selected outcomes

- Checkout: Rebuilt on Stripe Payment Element from six or seven to three steps. Payment conversion up 23%, above a 20% target; AutoPay opt-in roughly doubled.
- Card reader, 0 to 1: Relaunched on alpha fixes and grew to 17 practices. The card-on-file pilot cut post-visit statement costs 77%.
- AI support: Designed, 0 to 1, the customization for AI support chat and AI phone calls, with human review before a response is sent.
- Partner API docs: Turned a bare list of endpoints into documentation with field explanations, use cases, test calls, and a copy-for-LLM button. It opened partnership and integration conversations.
- Design system: Built in my first two months, then consolidated several scattered systems and the engineering library into one.
- Research in Guatemala: Shadowed 9 teammates across 14 sessions and logged 138 improvement opportunities, which led to forming the product operations team.
- Carrier LYNX: Excursion evaluation from one to five days to 15 minutes, per Carrier’s own materials.
- McDonald’s SMART Request: process 10 steps → 3; request time days → 17 minutes in early tests.
- The Home Depot AR room scanner: registrations up 40%, project completion rate up 76%, add-to-cart rate up 20%.

## Public routes

- [The system](/system): Current scope, operating model, and team learning.
- [About Jordan](/about): Leadership, teaching, and career context.
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.
- [Inbox Health Card Reader](/work/card-reader): 2025-2026 · Senior UX Designer → Director of Product Design → Director of Product Design & AI Operations · Relaunched after alpha fixes and grew to 17 practices.
- [Inbox Health AI Support](/work/ai-support): 2024-Current · Designed, 0 to 1, the customization for AI support chat and AI phone calls, with human review before a response is sent.
- [Inbox Health Partner API Docs](/work/partner-api): 2026 · Turned a bare list of endpoints into documentation partners can build on, with a copy-for-LLM button.
- [The Home Depot](/work/home-depot): 2020-2021 · UI/UX Designer → Senior UI/UX Designer · AR room scanning app that generates floor plans from a phone camera.
- [Hibernate](/work/hibernate): 2021-2022 · Design Lead, Brand Experience · Complete brand refresh: identity, app, packaging, and pitch deck.
- [Sight](/work/sight): 2019-2021 · UI/UX Designer → Senior UI/UX Designer · Dashboards for a machine-vision product that read the cameras a site already had.
- [Play](/play): Hands-on technical experiments and personal software.
- [Selected work](/work): 10 public case studies across healthcare, logistics, operations, and connected products.
- [Resume](/resume): Current role, selected outcomes, and leadership context.

## Portfolio-grounded question guide

### I’m a hands-on design leader. I build the product and the conditions around it.

Suggested question: How does Jordan work with Product and Engineering?

I’m Director of Product Design & AI Operations at Inbox Health, reporting to the CEO, and the sole product designer across a seven-app platform serving 2.5M+ patients. [2][4]

Hands-on product design, design engineering, and operating systems that turn conversations into reviewable work. Across the work, I make complex operational systems clearer through product judgment, hands-on craft, and reusable ways of working. [1][3]

The clearest fit is a hands-on Director, Staff, or Principal product-design role where research, systems, and building in code all count. [1][2]

Sources:
- [The system](/system): Current scope, operating model, and team learning.
- [About Jordan](/about): Leadership, teaching, and career context.
- [Selected work](/work): 10 public case studies across healthcare, logistics, operations, and connected products.
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.

### Start with Carrier LYNX and McDonald’s SMART Request.

Suggested question: What did Jordan personally do on Carrier LYNX?

On Carrier LYNX, I led the enterprise workflow design for a manual assessment process drawing on 10+ data sources. Carrier’s own materials put the automated assessment at 15 minutes, against the one to five days the manual version took. [1]

On McDonald’s SMART Request, I simplified an operational workflow and navigated the stakeholder work around it. The process moved from 10 steps → 3; in early tests, request time moved from days → 17 minutes. [2]

Together, those cases show hands-on product judgment and outcomes. My current Director scope adds the operating-system and leadership work around delivery. [3][4]

Sources:
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.
- [The system](/system): Current scope, operating model, and team learning.
- [About Jordan](/about): Leadership, teaching, and career context.

### At McDonald’s, I owned discovery, research, prototyping, and the workflow redesign.

Suggested question: How did Jordan work with McDonald’s executives?

I reduced a fragmented 10-step Excel process to a 3-step dashboard, mapped Red Routes, and translated the work into Epics, Stories, and Tasks for the development team. [1]

I worked directly with stakeholders, presented the prototype to McDonald’s global corporate leadership, incorporated their feedback, and ran additional edits before implementation handoff. [1]

The case study reports request time moving days → 17 minutes in early tests. That is an early-test result, not a documented long-term production measure. [1]

Sources:
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.

### At Inbox Health, I lead product design for patient-payment experiences and the systems around them.

Suggested question: What did Jordan personally do to increase payments?

I rebuilt patient checkout end to end on Stripe Payment Element, cutting the steps from six or seven to three. Payment conversion rose 23%, above a 20% target, and the redesigned checkout roughly doubled AutoPay opt-in. A controlled Stripe Elements test before the rebuild lifted the payment-method step +20.3% and successful payments +7.7%. [1]

I designed a 0-to-1 card reader end to end on the Stripe S700, ran the alpha on site in Maine, and relaunched it on the fixes; it grew to 17 practices. [2]

I also designed the customization for our AI support chat and AI phone calls, redesigned the partner API docs, and built and consolidated the design system. The platform-scale figures are context, not outcomes I claim as my own. [3][4][1]

Sources:
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.
- [Inbox Health Card Reader](/work/card-reader): 2025-2026 · Senior UX Designer → Director of Product Design → Director of Product Design & AI Operations · Relaunched after alpha fixes and grew to 17 practices.
- [Inbox Health AI Support](/work/ai-support): 2024-Current · Designed, 0 to 1, the customization for AI support chat and AI phone calls, with human review before a response is sent.
- [Inbox Health Partner API Docs](/work/partner-api): 2026 · Turned a bare list of endpoints into documentation partners can build on, with a copy-for-LLM button.

### For The Home Depot, I designed an AR room-scanning and floor-plan experience.

Suggested question: What other consumer-product work does Jordan show?

The mobile concept detects walls, doors, and windows, generates a floor plan, and connects measurements to the product catalog. [1]

I also designed product visualization in the customer’s space and remote connection to in-store associates. [1]

The Home Depot results were a 40% increase in registrations, a 76% increase in project completion rate, and a 20% increase in add-to-cart rate. [1]

Sources:
- [The Home Depot](/work/home-depot): 2020-2021 · UI/UX Designer → Senior UI/UX Designer · AR room scanning app that generates floor plans from a phone camera.

### For Hibernate, I led a week-long sprint that unified the brand and product experience.

Suggested question: What other consumer-product work does Jordan show?

I shaped the brand strategy from the founder’s story, refined the identity, redesigned the app, and introduced in-app upselling opportunities. [1]

I applied the system across the app, website, packaging, investor deck, and social touchpoints. [1]

The public case study documents a complete shipped system and founder praise, but it does not provide a measured premium-conversion result. [1]

Sources:
- [Hibernate](/work/hibernate): 2021-2022 · Design Lead, Brand Experience · Complete brand refresh: identity, app, packaging, and pitch deck.

### For Sight, VisionX’s machine-vision product, I designed the dashboards that turned camera detections into site insights.

Suggested question: What is Jordan’s strongest customer-facing AI evidence?

Sight ran computer vision on the security cameras a site already had. I designed the dashboards for the leadership teams reading it, and a drive-up and pickup dashboard for a McDonald’s pilot. [1]

The work includes a cross-platform design system for desktop, tablet, and mobile, designed around data-latency and site-scale constraints. [1]

The public case study supports interface and systems work around AI outputs. It does not establish that I designed the underlying ML model or provide a measured business outcome. [1]

Sources:
- [Sight](/work/sight): 2019-2021 · UI/UX Designer → Senior UI/UX Designer · Dashboards for a machine-vision product that read the cameras a site already had.

### The work spans healthcare, enterprise operations, consumer products, brand systems, and machine vision.

Suggested question: Which projects demonstrate Jordan’s level?

For The Home Depot, I designed an AR mobile app that scans a room, generates a floor plan, connects measurements to the product catalog, and supports remote connection to in-store associates. [1]

For Hibernate, I led a week-long sprint that unified the brand across the app, identity, packaging, pitch deck, website, and social touchpoints. [2]

For Sight, VisionX’s machine-vision product, I designed dashboards that turned camera detections into site insights, plus a cross-platform design system for desktop, tablet, and mobile. [3]

My current healthcare work at Inbox Health adds patient payments, provider workflows, hardware-integrated payments, and AI-assisted billing communications. [4]

Sources:
- [The Home Depot](/work/home-depot): 2020-2021 · UI/UX Designer → Senior UI/UX Designer · AR room scanning app that generates floor plans from a phone camera.
- [Hibernate](/work/hibernate): 2021-2022 · Design Lead, Brand Experience · Complete brand refresh: identity, app, packaging, and pitch deck.
- [Sight](/work/sight): 2019-2021 · UI/UX Designer → Senior UI/UX Designer · Dashboards for a machine-vision product that read the cameras a site already had.
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.

### On Carrier LYNX, I led the workflow and product-design work within a cross-company team.

Suggested question: Show a difficult Product and Engineering trade-off.

At Digital Surgeons I was Associate Creative Director, XD, then Design Lead, Brand Experience. I ran workshops and interviews and worked with AWS to map the process and identify steps we could remove. I extended Carrier’s existing Verge Design System for monitoring and evaluation. [1]

Carrier, AWS, and Digital Surgeons were partners. Stakeholder management was central to our consulting work. I collaborated with a third-party vendor that introduced the underlying technology partway through the project, and worked daily with Carrier’s PM and engineers to build the product. [1]

The 60-80% labor reduction and the 10 → 1 systems figure come from Carrier’s own solution materials. They are projections against Carrier’s $1M+ annual labor baseline, and I have no measured result for either. What I can claim is the design: I mapped seven human steps in the current process and designed a flow with four, and in testing 3 of 4 Amgen quality specialists finished their first evaluation unaided. [1]

Sources:
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.

### My work connects design decisions to time, cost, and payment outcomes.

Suggested question: What did Jordan personally do on the strongest project?

At Inbox Health, I rebuilt patient checkout on Stripe Payment Element, cutting the steps from six or seven to three. Payment conversion rose 23%, above a 20% target. The platform serves 2.5M+ patients; that scale figure is context, not an outcome I claim as my own. [1]

At Carrier LYNX, Carrier’s own materials put the automated assessment at 15 minutes. Carrier described the starting assessment as fully manual, drawing on 10+ data sources and taking one to five days. In testing, 3 of 4 Amgen quality specialists finished their first evaluation unaided, the fastest in 11 minutes 15 seconds. [2]

McDonald’s SMART Request reports a process change of 10 steps → 3 and request time of days → 17 minutes in early tests. [3]

The Home Depot AR room scanner saw a 40% increase in registrations, a 76% increase in project completion rate, and a 20% increase in add-to-cart rate. [4]

The public case studies do not document statistical significance, an independent audit, or long-term persistence for those figures. The Carrier figures come from Carrier’s own documents. The McDonald’s figures are early-test results. [1][2][3][4]

Sources:
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.
- [The Home Depot](/work/home-depot): 2020-2021 · UI/UX Designer → Senior UI/UX Designer · AR room scanning app that generates floor plans from a phone camera.

### The public case studies do not establish statistical significance, independent audit, or long-term persistence for those figures.

Suggested question: What measurable outcomes can you verify?

What they do report is a 23% rise in Inbox Health checkout conversion, above a 20% target, Carrier’s own documented 15 minutes automated assessment, and McDonald’s request time moving days → 17 minutes in early tests. [1][2][3]

The Carrier 15 minutes, labor, and systems figures are Carrier’s own documented targets and projections, so I do not present them as measured results; the Amgen usability test is the measured one, with four testers. McDonald’s is explicitly an early-test result. The public material does not provide sample sizes, confidence intervals, audit records, or later production measurements. [1][2][3]

Sources:
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.

### I create ROI by shortening workflows, lowering cost, and leaving teams with a repeatable way to ship.

Suggested question: What did Jordan personally do on the strongest project?

The clearest proof is in the work: Inbox Health checkout conversion rose 23%, above a 20% target; Carrier’s own materials put excursion evaluation at days → 15 minutes; request time moved days → 17 minutes in early McDonald’s tests. [1][2][3]

The team return is a repeatable operating model. I turn conversations into testable requirements, carry the same criteria through design and implementation, and leave reusable instructions for research, design, testing, and review. [4]

The exact return depends on the mandate and baseline. The pattern is concrete: less operational time, lower cost, and a clearer path from ambiguity to shipped work.

Sources:
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.
- [The system](/system): Current scope, operating model, and team learning.

### I turn conversations into shared requirements, reviewable work, and reusable systems.

Suggested question: What did Jordan personally change?

I start with what people said, turn it into testable requirements, and keep the same criteria through design, implementation, and review. [1]

At Carrier, stakeholder management was central to our consulting work. I collaborated with a third-party vendor that introduced the underlying technology partway through, and worked daily with Carrier’s PM and engineers to build the product. I conducted hands-on user research and worked with AWS to map individual process steps and identify where we could remove steps. [2]

At McDonald’s, I owned discovery, research, and prototyping, presented the prototype to global corporate leadership, incorporated their feedback, and worked through stakeholder resistance before implementation handoff. [3]

Sources:
- [The system](/system): Current scope, operating model, and team learning.
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.

### I build team capability through teaching, management, and a reusable operating model.

Suggested question: How does Jordan work with Product and Engineering?

I built reusable instructions for research, design, implementation, testing, and review. I taught the workflow to designers, then improved it through team use. [1]

I managed and coached designers, marketers, and an intern. At SoluTech I wrote job descriptions and hired, and at Inbox Health I’ve been part of hiring loops for a design intern and a product manager. I teach Web Design at Quinnipiac University, taught UI/UX and motion graphics at the University of New Haven from 2019–2024, and co-led an AIGA Connecticut volunteer board of 12+. [2]

I run design critique frequently, as I have in every design role and in my classes. I also mentor on ADPList. ADPList Top 1% Global Mentor, October 2022. 25+ mentorship sessions and 500+ mentorship minutes. [2]

Sources:
- [The system](/system): Current scope, operating model, and team learning.
- [About Jordan](/about): Leadership, teaching, and career context.

### Carrier LYNX is the clearest public proof of end-to-end product and systems craft.

Suggested question: What did Jordan personally do on Carrier LYNX?

I still do the work. I’m Director of Product Design & AI Operations at Inbox Health and the sole product designer across product, systems, and internal tools. [1]

Figma is my main design tool. In my first two months at Inbox Health, I built a design system from scratch in it. [2]

The case study connects research, workflow design, information architecture, prototyping, UI design, and design-system extension to a shipped enterprise outcome. [3]

McDonald’s SMART Request adds a second example of interaction simplification: a fragmented 10-step Excel workflow became one 3-step dashboard. [4]

That case also shows a fully wired Sketch prototype, Red Routes, information architecture, and Epics, Stories, and Tasks prepared for the development team. [4]

Sources:
- [The system](/system): Current scope, operating model, and team learning.
- [About Jordan](/about): Leadership, teaching, and career context.
- [Carrier LYNX](/work/carrier-lynx): 2021-2023 · Associate Creative Director, XD → Design Lead, Brand Experience · Evaluation days → 15 minutes, per Carrier’s own materials.
- [McDonald’s SMART Request](/work/mcdonalds-smart): 2019-2020 · UI/UX Designer · Process 10 steps → 3 · Request time days → 17 minutes in early tests.

### I design AI products and run the operating model behind them.

Suggested question: How does Jordan work with Product and Engineering?

Since June 2024 I’ve designed AI and conversational features at Inbox Health: LLM-suggested replies for patient-support agents, AI call transcription with real-time response suggestions, and per-client AI prompt configuration with simulation. [1]

I designed, 0 to 1, the customization for our AI support chat and AI phone calls. Practices set scripts, rules, escalation to a human, and hours; a side-by-side preview tests scenarios, including against a real patient record; support staff review a response before it is sent. [1]

I also designed an in-app AI support assistant for billers and practice staff, with starter questions, settings awareness, and escalation to a human agent, and I evaluate AI automations with blind replays scored on a parity scoreboard. [1]

I build in code myself: I ship TypeScript and React at Inbox Health with Claude Code, through a workflow that turns conversations into requirements and returns reviewable work. I have also built personal tools with React, TypeScript, Python, browser APIs, and Plaid. [2][3]

Sources:
- [Inbox Health AI Support](/work/ai-support): 2024-Current · Designed, 0 to 1, the customization for AI support chat and AI phone calls, with human review before a response is sent.
- [The system](/system): Current scope, operating model, and team learning.
- [Play](/play): Hands-on technical experiments and personal software.

### Copy the thread to continue the same evidence review in another LLM.

Suggested question: What should I ask Jordan in an interview?

The copied package includes the conversation, grounded answers, exact public source URLs, and an instruction to separate verified facts from inference. [1][2]

My machine-readable portfolio brief is also available directly. Neither surface includes expanded or private portfolio details. [1]

Sources:
- [The system](/system): Current scope, operating model, and team learning.
- [Selected work](/work): 10 public case studies across healthcare, logistics, operations, and connected products.

### Hire me where research, systems, and building in code all count.

Suggested question: What should I ask Jordan in an interview?

I’m the sole product designer across a seven-app platform, I report to the CEO, and I ship TypeScript and React myself. I authored the design organization’s 2026 OKRs and linked every sprint task to a product or design initiative. [1]

I’ve managed people at three companies: a designer, a contract designer, and two marketers at SoluTech, where I wrote job descriptions and hired; a junior designer at Digital Surgeons for about a year; and, briefly, a product designer, then the summer design intern, at Inbox Health. I co-led AIGA Connecticut’s volunteer board of 12+. [1]

The work on this site is the evidence: research on site, a design system consolidated across products, and features I built in code, each with its measured result where one exists. [2][3]

Sources:
- [About Jordan](/about): Leadership, teaching, and career context.
- [Selected work](/work): 10 public case studies across healthcare, logistics, operations, and connected products.
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.

### Good questions to ask me.

Suggested question: Which roles fit Jordan?

How did the checkout rebuild get from six or seven to three steps, and how was the 23% lift measured? [1]

What changed after running the card reader alpha on site in Maine? [2]

How do you keep AI output reviewable before it reaches a patient? [3]

Sources:
- [Inbox Health](/work/inbox-health): Checkout conversion up 23%, above a 20% target · 2.5M+ patients.
- [Inbox Health Card Reader](/work/card-reader): 2025-2026 · Senior UX Designer → Director of Product Design → Director of Product Design & AI Operations · Relaunched after alpha fixes and grew to 17 practices.
- [Inbox Health AI Support](/work/ai-support): 2024-Current · Designed, 0 to 1, the customization for AI support chat and AI phone calls, with human review before a response is sent.

### I cannot verify that here.

Suggested question: What kind of product-design leader is Jordan?

I can answer questions about my role, selected projects, measurable outcomes, leadership, craft, and way of working. I cannot verify compensation, availability, references, every contribution boundary, or private team details here. [1][2]

That boundary is intentional. Use the contact path on the About page for anything the portfolio cannot answer. [1]

Sources:
- [About Jordan](/about): Leadership, teaching, and career context.
- [Resume](/resume): Current role, selected outcomes, and leadership context.

## Use in another LLM

Use only the cited public evidence. Separate verified facts from interpretation. Preserve Jordan/team contribution boundaries. Do not invent private details, numbers, quotes, responsibilities, or outcomes. State clearly when the public portfolio cannot answer a question.