Phia AI Concierge: learns, styles and curates for you

Phia AI Concierge learns your style from Swipe Studio, wishlists, carts, mood boards, and photos, so users can chat with it for curated outfits, recommendations, and personalized discovery.

Comment

Inspiration

I wanted to make shopping feel more personal, interactive, and intelligent. Instead of endless browsing, I wanted users to have an AI fashion concierge that understands their style and helps them shop better.

What it does

Phia AI Concierge learns a user’s style from Swipe Studio, wishlist, cart, mood boards, and uploaded photos. Users can chat with it to get curated outfits, recommendations, trend suggestions, and style help. It also includes a Discovery tab for quick personalized suggestions.

How I built it

I built it by combining conversational AI with user behavior signals. Swipe Studio helps capture preferences, while wishlist, cart, mood boards, and photos help the AI understand the user more deeply and improve recommendations over time.

Challenges I ran into

The biggest challenge was making the AI feel truly personalized instead of generic. Another challenge was connecting all user interactions into one smooth experience.

Accomplishments that I’m proud of

I’m proud of building an AI concierge that does more than answer prompts, it learns from user behavior and curates fashion in a more personal way.

What I learned

I learned that strong personalization comes from combining multiple user signals, and that AI becomes much more useful when it understands user behavior over time.

What’s next for Phia AI Concierge: learns, styles and curates for you

Next, I want to make the AI even better at understanding changing style preferences, improving photo-based recommendations, and delivering smarter, more context-aware fashion suggestions.

Project Work Flow :- Phia AI Concierge is a personal shopping assistant that becomes smarter with every user interaction.

It starts by understanding the user through Swipe Studio, where users swipe on outfits and clothing items. From those swipes, the AI begins identifying the user’s vibe, aesthetic, preferences, and dislikes. But it does not stop there.

The AI Concierge continuously learns from the user’s wishlist, cart activity, and mood boards, building a stronger understanding of personal style over time. This allows it to deliver more curated recommendations, sharper outfit suggestions, and more personalized responses instead of generic shopping results.

Users can directly talk to the AI Concierge and ask it to:

  • curate outfits for a specific event
  • recommend pieces for a certain vibe
  • find options within a budget
  • suggest trending styles and accessories

Users can also upload photos to the AI Concierge, allowing it to understand visual preferences better and curate similar looks, matching outfit ideas, or style recommendations based on the uploaded image.

To make suggestions easy to access, Phia also includes a Discovery tab, where users can quickly explore AI-curated recommendations, trends, and personalized style picks in one place.

So instead of making users search endlessly, Phia AI Concierge first understands the user through swipes, wishlists, mood boards, cart behavior, and even uploaded photos, and then turns that understanding into intelligent, personalized fashion guidance.

Built With

Share this project:

Updates