CULTURAL_MATCH_AI

Transform your global expansion with AI-powered cultural insights. Understand local tastes, habits, and preferences to make your brand truly resonate worldwide.

Comment

Inspiration

Launching a product in a new market, opening a store in another city, or simply communicating with a new audience... it's always a challenge.
Not because there's a lack of tools. There are hundreds of them.
But because there's a lack of real understanding of the local culture.

Too many retailers, brands, and startups fail — not because of a bad product — but because they don't know what their audience values, likes, understands, or really expects.

  • ❌ A misinterpreted color
  • ❌ A reference that doesn't resonate
  • ❌ A product that doesn't fit local customs

In a world where every region, every city, and every neighborhood has its own codes, culture becomes a strategic factor.
And yet, there is currently no simple tool to help capture and understand these cultural nuances.

This is the gap that Cultural AI seeks to fill:
Offering an AI-based strategic assistant capable of guiding you by translating culture into concrete actions.
A tool that ensures every product, every idea, and every message truly speaks to its target audience.

How we built it

  • Frontend: React + TailwindCSS, with a 100% conversational interface
  • Serverless backend: Supabase Edge Functions
  • Generative AI: Gemini 1.5 / Gemini 2.5
  • Cultural data: Qloo Taste AI API

Challenges we ran into

The biggest challenge was orchestrating the dialogue between several external services — Qloo, Gemini, Supabase — while keeping the user experience smooth and fast.

Other difficulties included:

  • Adapting cultural insights to different contexts without falling into stereotypes or generalizations
  • Understanding the documentation and correctly using the Qloo API
  • Facing API limitations (especially request caps), due to using a free model often overloaded

🏆 Accomplishments we're proud of

  • Designing an app that can be used by everyone, even without technical skills
  • Successfully translating culture into actionable recommendations
  • Seamless integration of a specialized API (Qloo) into a modern AI pipeline

What we learned

  • Working with cultural data requires sensitivity and contextual understanding
  • AI is powerful, but only as good as the input and its alignment with real needs
  • A conversational interface lowers the barrier to entry, making complex tools feel simple

What's next for CULTURAL_MATCH_AI

We want to go further:

  • Localize the experience even more with hyper-contextualized suggestions
  • Integrate more LLM models for a smoother, smarter user experience
  • Add features for user accounts, personalized profiles, and custom LLM key management
  • Introduce a "cultural discovery" mode for curious users and travelers
  • Enable generation and export of reports in PDF and DOCX formats

Built With

Share this project:

Updates