Amanda Schmutzler
Marty
Walmart's generative AI partner agent, built into Seller Center so sellers and suppliers can ask how to do something and get an answer that links straight to the right tool.
- Problem: help lived in guides, Seller Academy, and support, and simple questions often ended up as tickets.
- Approach: one place to ask inside Seller Center, with answers that link to the right tool.
- Process: designed Marty with the Trust & Safety product team, from the Terms of Use to step-by-step answers.
- Outcome: sources and a thumbs up or down on every answer, so sellers can check what Marty says.
Introduction
Marty is Walmart's generative AI partner agent. It helps sellers and suppliers navigate tasks, explore tools, and grow their business directly within the platform. Marty sits in Seller Center: sellers ask it how to do something, and it points them to the right tool.
I designed the Marty experience with the Trust & Safety product team, from the Terms of Use sellers see before their first question to the numbered, step-by-step answers they get back.
My Role
I designed Marty as part of the Trust & Safety product team, alongside product and engineering. I designed the experience and the states a conversation moves through, from the Terms of Use to step-by-step answers, and handed off flows, answer layouts, and loading and error states to engineering.
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 placesGuides, Seller Academy, and support each held part of the answer, so no one place had all of it.
02 · Tools were spread outSellers had to already know where to go in Seller Center to find the tool they needed.
03 · Questions became ticketsSimple questions often went to support instead of being answered where the seller was working.
How Marty works
Four steps take a seller from a question to an answer they can check.
Ask a question
Sellers type it the way they would ask a person.
Wait a moment
A “Thinking…” state shows Marty is working on it.
Get an answer
With links to the Seller Center pages they need.
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 AITerms of Use before first use, and a short reminder under every chat box.
Don’t start with a blank boxSample questions and topic buttons on the welcome screen.
Show it’s workingA “Thinking…” state shows sellers it’s still working.
Link to the toolAnswers link to pages like Media Library and Add Items.
Allow follow-upsAsk “walk me through the steps” and get a numbered list.
Let sellers push backThumbs 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 in use
A question moves through four states. Each one tells the seller what Marty is doing and what they can do next.
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 I learned
I learned that an AI answer earns trust by showing its work. Sources one click away and a thumbs up or down on every answer gave sellers a way to check and push back, which matters more than making an answer sound sure.
I learned that the blank box is the hardest moment. Sample questions and topic buttons on the welcome screen gave sellers somewhere to start, instead of making them guess what Marty could do.
I learned that being upfront about AI belongs in the interface, not only in the fine print. The Terms of Use before first use and the reminder under every chat box set expectations before anyone relied on an answer.
Why this kind of problem fits me
Marty is the kind of problem I look for: a new technology that has to feel clear and honest to the people using it every day.
I like designing for the moment someone needs one clear next step, not another place to search.
AI is new, but sellers want a clear answer, so I turn it into states and answers they already understand.
I like work where honesty shows up in the interface: what Marty is, what it's doing, and where answers come from.
Why Bring Me Onto Your Team
I've designed an AI assistant's flows, states, and answers with a product team, and I use AI in my own design work every day, so I know what it looks like when it helps and when it misses. I design AI that is upfront about what it is, easy to check, and easy to push back on.
If your team is bringing an assistant to the people who use your product, I'd like to help make it clear and trustworthy.
