AI Agents and Creating Chat Bots

How do you get people to trust an AI? I learned the rules designing Marty at Walmart, then used them to build and train Mandy, my own assistant, and a guest chatbot I designed and built for Vegas Suite Rentals.

RolePrincipal UX Designer on Marty (Walmart), designer, builder, and trainer of Mandy and the Vegas Suite Rentals chatbot
IndustryGenerative AI, AI agents, chatbots
Impact3chatbots designed24/7answers, any hour4trust rules, used in all three
At a glance
  • Problem: people don't know what an AI assistant can do, or whether to trust what it says.
  • Approach: learn the rules of trust on a big product, then test them on assistants I build myself.
  • Story: Marty at Walmart taught me four rules. I used them to design, build, and train Mandy. Then I used them again for a Vegas Suite Rentals chatbot.
  • Outcome: Marty gives sellers answers they can check. Mandy answers in my words and never sends an offer without your approval. The Vegas chatbot answers guests in the owners' voice at any hour.

Introduction

AI assistants are easy to build and hard to trust. The technology can answer almost anything, which is exactly the problem: people don't know what it can do, whether it's right, or who is in charge when it isn't.

This is the story of how I learned to design for that, in three steps. At Walmart I helped design Marty, a generative AI partner agent in Seller Center, and came away with four rules. Then I used those rules to build Mandy, the assistant on this site, from interviews of my own work. Finally I used them for a chatbot for Vegas Suite Rentals that answers guests in the owners' voice. Each one taught me something the last one couldn't.

Designing an AI conversation

My Role

On Marty, I designed the experience and the states a conversation moves through, alongside product and engineering. On Mandy and the Vegas chatbot, I did all of it: the idea, the voice, the interface, the questions and answers, the links to my work, and the build. I connected both to Claude, Anthropic's AI model, and trained them by writing, testing, and rewriting their answers.

Part one: Marty

Where I learned the rules.

Marty sits in Seller Center. Sellers ask how to do something and get an answer that links to the right tool, with sources one click away and a thumbs up or down on every answer. I helped shape what Marty says and how it behaves, including researching what sellers ask and helping write the prompts that train it.

Working on Marty showed me what makes an assistant easy to trust, and it is mostly not the AI. It is what surrounds the answers: telling people it's AI before they start, and letting them check and push back on every answer.

Where sellers got stuck

Answers existed, but they were spread across Seller Center, and finding the right one took a seller who already knew where to look.

01 · Help lived in a lot of places

Guides, Seller Academy, and support each held part of the answer, so no one place had all of it.

02 · Tools were spread out

Sellers had to already know where to go in Seller Center to find the tool they needed.

03 · Questions became tickets

Simple questions often went to support instead of being answered where the seller was working.

Research and prompts

A chatbot is only as good as what it is taught to say. Working closely with engineering, I did a lot of research into what sellers ask and where the answers live, and I helped write the prompts that train the chatbots.

What I researched

Seller questions and tickets

I looked at the questions sellers were asking and the support tickets they turned into, to help the team see which problems came up most and how sellers phrased them in their own words.

The existing help content

I went through Seller Center, Seller Academy, and the guides to help map where each answer lived, where the content overlapped, and where it was missing, since Marty could only be as complete as the sources behind it.

Conversations with people

I talked with the people closest to the problem, including sellers, support, and subject-matter experts, to help the team understand what a good answer looks like and what a wrong one costs.

Other chatbots and AI best practices

I looked at how other chatbots handle answers, sources, and mistakes, and what current AI best practices say about being clear on what the bot can and cannot do, and brought that back to the team.

What the prompts covered

Tone and voice

I helped set how Marty sounds. Plain, helpful, and in the language sellers use, not the language of internal documentation.

Answer structure

I helped shape the answers themselves: short responses, numbered steps when a seller asks to be walked through something, and links to the Seller Center tool that finishes the job.

Guardrails

I helped define what Marty should decline, when it should send a seller to support instead of guessing, and how it handles uncertainty, so a wrong answer never looked like a confident one.

Testing and iteration

Alongside engineering, I helped run sample questions against the prompts and rewrite them based on what came back, repeating until the answers held up across the kinds of questions sellers actually ask.

How Marty works

Four steps take a seller from a question to an answer they can check.

01

Ask a question

Sellers type it the way they would ask a person.

02

Wait a moment

A “Thinking…” state shows Marty is working on it.

03

Get an answer

With links to the Seller Center pages they need.

04

Check and rate it

Sources are one click away, plus a thumbs up or down.

Design decisions

Six decisions shaped how Marty behaves, from what sellers see before their first question to what they can do with the last answer.

Be upfront about AI

Terms of Use before first use, and a short reminder under every chat box.

Don’t start with a blank box

Sample questions and topic buttons on the welcome screen.

Show it’s working

A “Thinking…” state shows sellers it’s still working.

Link to the tool

Answers link to pages like Media Library and Add Items.

Allow follow-ups

Ask “walk me through the steps” and get a numbered list.

Let sellers push back

Thumbs up or down and sources on every answer.

Being upfront about AI

Before a seller asks anything, Marty asks them to read and accept the Terms of Use. After that, a short reminder under every chat box says that AI can make mistakes and not to share sensitive information.

Marty Terms of Use shown before first use
Accept terms before startingSellers review and accept the Terms of Use before asking Marty a question.
AI can make mistakesA reminder beneath the chat tells sellers that AI responses can contain mistakes.
Protect sensitive informationSellers are reminded not to share sensitive information in chat.

Marty in use

Here is one question from start to finish. Each step tells the seller what Marty is doing and what they can do next. It plays on its own, and you can step through it yourself.

Walmart Seller Center
Marty screen

    What I handed to engineering

    I handed off the flows for each state, so engineering knew what to build and how each answer should behave.

    What I handed off

    • Flows for each state
    • Answer layouts
    • Loading and error states

    What I specified

    • Sources on every answer
    • Thumbs up or down
    • Follow-up questions

    What engineering would build

    • Answers from Walmart data
    • Links and sources
    • Feedback buttons

    What shipped

    Marty gives sellers and suppliers one place in Seller Center to ask how to do something, with answers that link to the right tool, sources one click away, and a thumbs up or down on every answer. Simple how-to questions get answered where the seller is already working, instead of becoming support tickets.

    I contributed to the design alongside product and engineering, from the Terms of Use through the loading, error, and step-by-step states.

    What Marty taught me

    Four rules came out of Marty. They are the rules I carried into everything after it.

    Rule 1 · Say what it is

    People should know they are talking to AI, and that it can be wrong, before they rely on an answer.

    Rule 2 · Show its work

    A wrong answer that sounds sure is worse than no answer. Sources and links let people check.

    Rule 3 · Keep a person in charge

    People can rate an answer, push back, or reach a human. The assistant never has the last word.

    Rule 4 · Stay in scope

    An assistant that knows what it can answer, and where to send everything else, is easier to trust than one that tries everything.

    Those four rules are what I took with me. The next question was whether I could apply them on my own, from start to finish.

    Part two: Mandy

    Where I used the rules on myself.

    Mandy lives on this site. Visitors can ask her about my work, background, and process, and she answers in short, linked bullets, at any hour.

    I wanted a real, working example of how I design AI, not just a description of it. Marty had a large team and a long review process. This time there was no team and no review. Every decision, and every wrong answer, was mine.

    Try her, and the hiring agent

    This is the real assistant. Ask her something, or switch to the hiring agent: tell it your company and the role, and it researches, drafts three reasons you need the role, reads my recommendations, and prepares an offer. Then it stops and waits for you to approve or decline. It is scripted on purpose, so it never does anything real.

    Nothing you type is saved.

    Mandy

    Designing Mandy

    The tags show which of Marty’s four rules each decision came from, and which are new from using her.

    Rule 1 · Say what it is

    She introduces herself as my AI, and says I designed, built, and trained her, before anyone asks.

    New · No blank box

    Starter questions and topic buttons on the welcome screen, so the first question is a tap.

    New · Short, bulleted answers

    A lead-in and a few bullets are quick to scan. Walls of text are not.

    Rule 2 · Show its work

    Answers point to the page and section on this site, so a visitor can check the work themselves.

    New · Start fresh every time

    Closing the chat clears it. Nothing a visitor types is kept.

    Rule 3 · Keep a person in charge

    "Contact Amanda" stays at the bottom of the chat for anything she should not answer.

    How I made Mandy

    Design only gets you the shell. What Mandy says had to come from me, so I made her by interviewing myself, then turning those answers into an assistant I trust to speak for me.

    01

    Record interviews

    I recorded interviews of myself answering the questions a recruiter or hiring manager would ask, in my own words.

    02

    Import the transcripts

    I imported the transcripts and the questions into the agent, so her answers start from how I really talk about my work.

    03

    Hook it up to Claude

    I connected the agent to Claude, with a small server file that keeps the AI key off the page.

    04

    Test and rewrite

    I asked her questions, read what came back, and changed the answers wherever they missed, again and again.

    What training an agent was like

    Training Mandy felt less like coding and more like coaching a new hire who has read my whole career but has never met me. The interviews gave her my words. Everything after that was editing: deciding what good looks like, then saying it clearly enough that she does it every time. It took a lot of hands-on work to make it feel right.

    Facts only

    She answers from what I told her in the interviews. She is told never to invent employers, dates, metrics, tools, or opinions, because a made-up detail about my career would do real harm.

    Answers that show the work

    I linked images and projects to questions, so an answer about a project can point to the exact page and show what I made, not only describe it.

    A voice, and a format

    Every answer opens with a short lead-in, then two to five short bullets, one idea each. She is warm and professional, and she sounds like one assistant every time.

    Rule 4 · Stay in scope

    She stays on my work, background, process, and how to reach me, and politely declines the rest. She does not share confidential details from past employers, and she does not discuss salary or whether I would accept an offer.

    The hardest part was editing

    The early answers were too long and repeated themselves. Fixing that was most of the job. I would ask a question, read the answer like a critique, decide what was wrong, and change it: never write a paragraph, never repeat yourself, never ask more than one question. Each fix is small, but together they are the product.

    It is the same skill as design critique. You look at what came back, name what is off, and say what you want instead. The difference is that the assistant does the same thing every time, so a small, clear fix improves every future answer.

    Keeping her safe to run

    An assistant that speaks for me on a public site needs limits. The AI key lives in a separate server file, never in the page. Mandy only answers requests that come from this site, and there is a daily cap on messages so cost can't run away. She runs on Claude Haiku, a small, fast model, which suits short, quick answers.

    Part three: Pat & Bernie

    A chatbot I designed and built for two suite owners.

    Guests of Vegas Suite Rentals ask the same questions again and again: parking, check-in, pets, fees. I built a chatbot on the guest site that answers them at any hour.

    It introduces itself as "Pat & Bernie, Superhosts on Airbnb," answers in the owners' own voice, and is styled to match the dark and gold look of the site. Read the Vegas Suite Rentals case study.

    Vegas Suite Rentals chatbot

      One set of rules, three assistants

      The four rules I learned on Marty held up every time, whether the assistant was for a large company, for me, or for two suite owners.

      Rule 1 · Say what it is

      Marty shows Terms of Use and an AI reminder before the first question. Mandy introduces herself as AI in her first line. The Vegas chatbot introduces itself up front and invites the first question.

      Rule 2 · Show its work

      Marty puts sources one click away. Mandy links to the exact spot, and the hiring agent shows every step it takes.

      Rule 3 · Keep a person in charge

      Marty lets sellers rate answers and push back. Mandy asks you to approve the offer and always offers a way to contact me.

      Rule 4 · Stay in scope

      Marty has guardrails for what to decline. Mandy only talks about my work. The Vegas chatbot sticks to questions about a stay.

      What I learned

      Looking back, I learned that trust is a design problem. Saying what the assistant is, showing where answers come from, and keeping a person in charge did more for trust than making the assistant sound confident.

      I learned that training an agent starts with real material and ends with editing. Interviewing myself gave Mandy my words, and the work after that was reading her answers closely and rewriting the ones that missed.

      I learned that agents need a visible plan and a human decision. The hiring agent works because you can see each step and nothing happens until you say so. That is the same idea behind Marty's sources and thumbs up or down.

      What I learned about trustworthy AI

      Why this kind of work fits me

      Assistants and agents are the kind of problem I look for. A new technology that has to feel clear and honest to the people using it.

      01 · Helping someone who is stuck

      I like designing for the moment someone needs one clear next step, not another place to search.

      02 · New technology, plain behavior

      AI is new, but sellers want a clear answer, so I turn it into states and answers they already understand.

      03 · Trust built into the design

      I like work where honesty shows up in the interface, like what Marty is, what it's doing, and where its answers come from.

      Why Bring Me Onto Your Team

      I designed the experience for Walmart’s AI assistant, then used what I learned to design, build, and train two more of my own. I know how to make an assistant clear, honest, and safe in front of real people, and I use AI in my own design work every day.

      If your team is bringing an assistant or an agent to the people who use your product, I'd like to help make it trustworthy.

      Get in touch

      Have a hard problem worth designing through? Let's talk.