VibeMatch AI
Customizing AI Fashion Models Based On User-Uploaded Images​
As a product designer at Huhu.AI, an AI fashion try-on startup, I designed and delivered a new feature—customizable AI fashion models that vibe-match users' uploaded images—in just 1.5 weeks. I worked closely with the co-founders and senior AIGC scientists throughout the process.
This feature is crucial in securing Huhu.AI’s first-ever business deal, a partnership with Cider.
ROLE
Product Designer
Prompt Engineer
TEAM
1 Product Manager, 1 Front-End Engineer, 1 Back-End Engineer, 1 AIGC Scientist
DURATION
Oct 28th - Nov 8th, 2024 (1.5 Weeks)
WHAT I DID
Prompt Engineering
AI Research
Wireframing
Prototyping
Design Review Workshop
AI Template Crafting
CONTEXT
Huhu AI is a platform to create custom AI models and generate try-on images.
The service creates AI models with user-selected traits (e.g., gender, race, age) and generates try-on images for these or user's own models, targeting small to medium-sized businesses to save photoshoot costs.
Service 1: Fully Customizable AI models
Service 2: AI Try-on Images
PROBLEM
But, simply modifying the pre-defined attributes is limited in capturing the unique "vibe" of the model our clients envision.
Sometimes, our clients come with a specific mood board in mind that cannot be fully captured by adjusting predefined attributes. To address this, we need a feature that allows users to generate an AI model image based on a reference image of their choice. This feature should offer control over the level of similarity to the reference, as well as flexibility in adjusting other attributes like clothing, scene, and pose.
PROCESS
I delivered the feature design within 1.5 weeks, leading everything from interface design to parameter adjustments and running prompt engineering experiments.
Due to NDA, please reach out for more behind-the-scene process.
IMPACT
This feature played a key role in helping Huhu.AI secure its first business deal following its pivot in August 2024.
REFLECTION
Experimentation is Key to Human-Centered AI
Joining this project just weeks after starting was an exciting journey. By diving deep into our AI model, I gained a solid understanding of its capabilities and limitations, which became the foundation for designing intuitive interfaces.
Navigating ambiguity while running prompt engineering experiments was both challenging and rewarding. By iteratively testing and refining keywords, I uncovered patterns in the model's behavior, which informed both the design and functionality of the feature.
Efficiency Comes With Good Communication
A 1.5-week timeline required intense focus and adaptability. I ensured continuous alignment with the team to keep us moving efficiently.
Usability Over MVP Constraints
Even in the rush to launch, usability remained a priority from my perspective. By prototyping multiple interface options and hosting a design review, I ensured the solution worked seamlessly for users while finding practical compromises with developers.