AI · 2026
Axon
An AI workflow automation web app that asks a person when it is unsure
AI case studies: products where the AI does real work for the user
Seven AI product design case studies, from an app that redesigns a room from one photo to AI agents that run influencer campaigns. In each one the AI does a real job for the user, and the design makes that job clear and easy to check.
- case studies
- 8
Home and interior · 2026
RoomMind AI
An iOS app that redesigns a room from one photo
Fintech · 2025
Nexfolio
An AI-powered investment dashboard
Real estate · 2026
Lyvio
Website and product screens for an AI property advisor
AI · 2024
AI analysis website
A landing page for an AI business analysis product
Healthcare · 2026
CareUp
An AI healthcare app for finding a doctor and booking
Fintech · 2026
WealthWise
An AI personal finance dashboard
Marketing · 2026
Markly
An AI influencer marketing platform
What these AI projects have in common
An AI feature is only useful when people understand what it did and can trust it. So the design work here is less about the model and more about the conversation around it: what the person asks for, what they get back, how they can adjust it, and how they know when to double check. Results are shown next to the inputs that produced them, and the person stays in control of anything that matters.
The AI projects
- RoomMind AI is an iOS app that redesigns a room from one photo: pick a style and get an AI redesign in under 30 seconds.
- Nexfolio is an AI-powered investment dashboard with market forecasts and AI insights beside the portfolio.
- WealthWise is an AI personal finance dashboard with AI insights alongside cash flow, investments and goals.
- Lyvio is an AI property advisor that learns how a person lives, then recommends homes that fit.
- CareUp is a healthcare app with an AI health assistant that helps people find the right doctor.
- Markly is an influencer marketing platform where AI agents help brands discover creators, negotiate and manage campaigns.
- The AI analysis landing page sells an AI product that helps businesses analyse and optimise how they run.
How I approach AI product design
I design the AI as one part of a product, not as a chat box bolted on. First the job: what the AI does better than the person would alone. Then the moments around it: the input (a photo, a goal, a brief), the wait, the result, and the way back if the result is wrong. On RoomMind that is a style picker, a short wait and a before and after. On Markly it is an agent whose suggestions a brand can accept, change or reject. For more, see my AI product design page.
Frequently asked questions
It shows what the AI actually does for the user, the inputs and outputs, and how the interface handles waiting, uncertainty and mistakes. A screen with an "AI" badge is not enough: the case study should show the full flow from request to result.
Yes. Markly uses AI agents to discover creators and negotiate with them, and CareUp has an AI health assistant. Designing for agents means showing what the agent is doing, what it suggests, and where the person approves or steps in.
Yes. I build web front ends in React and Next.js and mobile apps in React Native or Flutter, and I design the states an AI feature needs, so the build covers waiting, errors and edge cases. A project starts with a free 30 minute intro call, then a written scope and one fixed quote.