AI-X-Artisans

Bridging the Digital Divide with Google AI


Inspiration

Small-scale artisans and traditional craftsmen create breathtaking, one-of-a-kind work, but they often struggle to survive in the digital-first economy. Professional product photography, SEO-optimized descriptions, and global marketing are expensive and technically daunting. AI-X-Artisans was built to provide every local creator with a world-class marketing team and a studio-grade digital storefront, ensuring that ancient traditions can thrive in a modern marketplace.

What it does

AI-X-Artisans is an all-in-one AI-powered commerce assistant that handles the "heavy lifting" of digital sales for artisans:

  • AI Photo Studio: Transforms raw, handheld mobile photos into professional catalog-ready shots using advanced style transfer and image enhancement.
  • Voice-to-Listing: Allows artisans who may not be tech-savvy to simply speak about their product. The AI extracts structured data (name, tradition, materials, price) and builds a complete listing in seconds.
  • Heritage Storytelling: Generates culturally accurate "Heritage Stories" for every product, embedded in Digital Authenticity Certificates with QR codes.
  • AI Negotiation Advisor: Helps artisans navigate price negotiations with customers, ensuring they earn a fair wage while remaining competitive.
  • Multi-Role Ecosystem: Connects artisans with Volunteers (who offer marketing/design skills) and Customers (who seek verified, authentic handcrafted goods).

How we built it

We built AI-X-Artisans using a modern, scalable stack centered on Google AI:

  • Frontend: Next.js (App Router) with Tailwind CSS for a premium, responsive UI.
  • Backend & Database: Supabase (PostgreSQL) for real-time chat, authentication, and secure file storage.
  • Google AI Engine (Backend):
    • Gemini 3 Flash: Voice-to-Listing structured extraction (JSON), visual price appraisal, and negotiation advice.
    • Gemini 3 Pro: Deep reasoning for culturally accurate Heritage Stories.
    • Vertex AI: Video generation pipeline blueprint and deployment-ready path for Veo integration.
    • Vision & Voice: Web Speech API for capture, Gemini 3 for multimodal understanding and structured output.
  • Architecture: Prisma ORM for type-safe database management and a middleware-based role switching system for the demo environment.

Challenges we ran into

  • Maintaining Texture Authenticity: It was critical that the AI didn't "over-process" handmade goods. We fine-tuned prompts to ensure unique textures (like handloom weaves) were preserved, not smoothed over.
  • Multilingual Support: Creating a seamless experience that works across languages (English/Hindi) while maintaining the nuance of traditional craft terms.
  • Agentic Extraction: Turning unstructured voice input into valid, schema-compliant JSON required rigorous prompt engineering and error handling.

Accomplishments that we're proud of

  • End-to-End Workflow: Successfully reduced the time to create a professional listing from 30+ minutes to under 30 seconds.
  • User-Centric Design: Creating a "Warm AI" experience that feels like a helper to the artisan rather than a complex technical tool.
  • Authenticity Verification: Building a system where consumers can actually verify the roots of what they buy, connecting the buyer to the maker.

What we learned

  • Accessibility is Key: For local artisans, multimodal input (voice/vision) isn't just a "feature"—it's an essential accessibility requirement for digital inclusion.
  • Hybrid AI Models: We learned that combining the speed of Gemini Flash with the depth of Gemini Pro creates the best user experience.

What's next for AI-X-Artisans

  • Hyper-Local Language Support: Expanding beyond Hindi/English to regional dialects.
  • Real-time Voice Negotiator: An audio-based agent that can handle live price negotiations between artisans and customers.
  • Supply Chain Transparency: Integration with blockchain to track the raw material journey for 100% ethical sourcing proof.

Built with passion to preserve heritage through innovation.

Built With

Share this project:

Updates