Carousell
Ongoing AI design works
Heads-up: This is not a full-fledged case study, but my ongoing notes on my AI design works done at Carousell. This is a living document.
I was the first designer at Carousell to explore and shipped AI features. I continue to explore and learn with other designers in the team.
Role
A learning designer 🤓
Timeline
Jan 2024 to present
List with AI
Time to list reduces by half, from 4 to 2 mins, after we released Multimodal Sell Form. I worked hands-on and mentored a senior designer on this project
Multimodal interaction
Stream of thought feedback
Shop with AI
I explored different ways AI can help buyers in shopping. Beside a chat agent, there are other useful design patterns. Buyer behaviour in each category is different, so is the most suitable AI interaction in each category.
Some projects are released, some are in exploration, waiting for development.
Visual category: Fashion and furnitureA low-effort browse interaction encourages visual discovery
structured, complex: AutosA guided and flexible intent refinement interaction
Unstructured and high knowledge gap: Home servicesChat agent helps buyers eliminate research phase to diagnose problems, find and contact sellers and get quotes.
AI design guidelines for Carousell
I initiated creating the guidelines and worked with a senior designer to define emerging patterns on our app. We aim to cover: product principles, tone of voice, interaction patterns and design system elements.
Go to next project