Product designer who ships
I design AI that
people actually trust.
I make AI for people who run real businesses and never wanted to run software, the kind that answers their customers over text, voice, and email, and I ship it to production myself in React. Talk to one on the right.
Designed and built end to end · React · Next.js · Claude
4 operators live1 product spine18 yrs shipping6 show network
C
Cedar & Oak Plumbing & HVAC
answers 24/7, books the job, hands off to a human when it should
online
Hey, you reached Cedar and Oak's front desk. I can answer questions, book a visit, or get a human on it. What's going on?
How I design AI people trust
Six rules I build by.
AI is easy to demo and hard to trust. These are the principles I design to, and under each is where I have already proven it in production.
01 An AI should refuse to guess.
If it cannot ground an answer, it says so. Confident wrong answers are how trust dies.
Proof: Sophia declines rather than fabricate.
02 Every claim needs a receipt.
The output traces back to a real, scored source, so a professional can stand behind it.
Proof: the citation-trust layer.
03 Human in the loop, not on cleanup.
Approval and handoff live inside the flow, before anything goes out, never as damage control after.
Proof: the approval gate, and the demo handing off to a person.
04 Design for the back of the room.
Under pressure, legibility beats decoration. I cut a tilde once because it read as a minus from across a room.
Proof: the live auction board.
05 Make the slow thing instant.
Find the step that kills momentum and collapse it. Speed is a feeling, and the feeling is trust.
Proof: a new combination, three minutes on a spreadsheet, now seconds.
06 The bar is zero hesitation.
Not a pretty mock. A person who never wanted software, using the thing without thinking about it.
Proof: operators adopting it with real money on the line.
The Desk · company-wide AI
One place that brings AI to everyone, and meets each person where they are.
The Desk is my own idea, and the work I am proudest of. One workspace that unites every tool a company already uses and puts an AI inside it. It learns each person's comfort level and adapts: more guidance for someone who never wanted software, out of the way for a power user, and it grows with them as they get more confident. The point is to bring AI to everyone on day one, not just the technical few, and lift what each person can do.
React · Next.js · multi-tenant · role-aware · adaptive UI · Claude
At one company, the whole team runs on it, every department, every day.
Good morning, Dana.
New here? I will walk you through it.
✦ Ask the Desk anything...or pick one below
Show me today's prioritiesDraft the funder updateWho is waiting on me?
✦ I pulled your three open tasks into a simple plan and drafted the funder update from Monday's notes. Want to review it together?
Review draftNot now
Tasks 3 due today
Docs 2 shared with you
Try it: switch Guided and Pro to watch the Desk meet a brand-new user, then a power user.
“Having the desk available is like working next to the most patient, knowledgeable, and experienced clerk-treasurers in the state.”City Clerk-Treasurer
Conversational AI · at scale
An AI that talks to thousands of people, and never makes things up.
I designed and built Sophia, the AI concierge for a national true-crime event owned by a major media company. She answers thousands of attendees over chat in the event's own voice, drawing on 600+ event sessions and 400+ verified, fact-checked speaker profiles, and she gets sharper as new transcripts land. The hard part was never getting her to answer. It was making her refuse to invent a fact she could not ground, so a brand could put her in front of its whole audience without flinching.
Next.js · Claude · retrieval grounding · guardrails · self-improving · live in production
A brand put it in front of its whole audience without flinching.
Sophia
AI concierge · knows the whole event
When do doors open Saturday, and is parking included?
Doors open at 9am Saturday. Parking is not included in the badge, but the host hotel has on-site self-parking. I can point you to the rate if you tell me your hotel.answer grounded in event FAQ
Who is speaking Sunday night?
The Sunday lineup has not been announced yet, so I would not want to guess. I will have it the moment it is public.declines rather than fabricate
grounded answers refuses to fabricate on-brand voice improves over time
Across the county, demand for infant care outpaces licensed capacity by a wide margin, with the sharpest gap in rural ZIP codes where the nearest provider can be over 20 minutes away.
1 County licensing roster, 2024 · capacity by age 98%
2 Stakeholder interviews, n=31 · access themes 94%
AL
A human reviews every claim before it ships
approve ✓
AI trust · human in the loop
AI a professional will put their name on.
For a consulting firm whose reports get read by funders, I designed a citation-trust layer. Every claim the AI writes traces back to the client's own source, scored, and a human approves it inside the flow instead of after. Making AI feel controllable and accountable is the whole job, and it is the same job here.
Next.js · retrieval · source scoring · approval gate
Consultants put their name on what the AI produced.
Real-time · dense data, made calm
A live land auction, resolved on screen.
Farmland sells live, by the parcel, while bidders sit in the room. People bid on a single tract, a combination, or the whole farm, and the board recomputes the optimal combination in real time. A new combination bid used to take two to three minutes to work out on a spreadsheet, and the room would lose its energy. Now the room sees the new first, second, and third in seconds. I designed it to be read from across the room with real money on the line, down to dropping a tilde that read as a minus sign from the back.
React · Tailwind · real-time bid-combination engine (dynamic programming)
The Desk in production
The same Desk, four very different seats.
The Desk above runs for operators who share almost nothing: a large farm-management company that put its whole team on it, a county office, a city clerk, and a consulting firm. One product and one AI, with row-level isolation, reshaped into a different job in every seat. That is the role-aware, many-organizations-on-one-product problem, solved and in daily use.
Four operators, one product, all live and used daily.
Farm management
A whole company, every department, on one Desk
A large farm-management company put its entire team on the Desk: leadership, accounting, marketing, and farm managers. AI grounded in the firm's own knowledge, an email assistant, and live numbers from the back office.
Across the company 35 on it
Ask the Deskchat
Email assistantdraft
Bank feed, livesynced
County government
A desk for an office the public is watching
A county official with zero prior AI experience ran a full 8-hour workday on it, drafting real policy, a contract review, and an RFQ, with no software staff in the building.
Today's queue live
Meeting minutesdraft
Grant finder3 new
Claims reviewdone
City clerk
Same Desk, a very different seat
Council minutes, ordinances, and public-records requests on the same product, reshaped for a city clerk's job.
Council, this week
Council minutesdraft
Ordinance 24-09posted
Records requestAPRA
Nonprofit consulting
The trust layer, in daily use
The citation-trust workflow above, lived in every day by a team that stakes its name on the output.
Claims, with receipts
Capacity gap, rural98%
Access, n=3194%
Human approvesapprove
0
operators live in production
0
product spine, role-aware
0
years shipping software
0
show network, built solo
Eighteen years of building
I have been shipping software to fit real work since 2008.
I taught myself to code to launch the first true-crime podcast, then kept building whatever the work needed. A sampling, beyond the AI products above.
Tools for myself
An ADHD-shaped project manager
I built my own task system around how my brain actually works, because the off-the-shelf ones fought me.
Consumer app
A podcast listening app
A clean way for an audience to listen, built when the tools to host and play audio barely existed.
Media network
A six-show podcast network
Brand, product, audience, and operations for an independent network, owned end to end.
The problem I want to work on
Earning a stranger's trust at 9pm.
The hard part of letting an AI answer a business's customers is not the answering. It is trust: an AI that earns it from a stranger after hours, handles the conversation that goes sideways, knows the exact moment to hand a person the wheel, and sounds like the business instead of a bot. The owner is putting their name behind it. Get the trust right and the growth follows. That trust surface is the work I want to own.
How I work on a team
I move at the speed of a team and raise the bar.
I own the problem, not just the screens. Discovery, the interaction model, the design system, the production code. Accountable for the outcome, end to end.
I work as a two, not a one. My co-founder runs discovery and the client relationship; I turn her real operator interviews into the shipped product, and every build is reviewed against her field notes.
I design from real use, not taste. After watching an operator's first session reach for an attach button that did not exist, I shipped in-chat upload, a clean copy, and document export next. The bar is the moment someone stops hesitating.
I make PMs and engineers faster. Working prototypes beat decks. I run critique, give and take feedback hard, and leave the codebase better than I found it.
Let's talk
I design the AI, and I ship the app.
If you are building AI that real people have to trust, that is the work I want to be doing.