Artifex

AI Repair Assistant

Comment 1

🔧 Artifex | AI-Powered Field Repair Assistant

Live Demo Powered by Gemini

Transform a single damage photo into professional repair documentation in 30 seconds

🎯 The Problem

Field technicians, repair shops, and students in Gombe face critical challenges:

  • No access to up-to-date repair manuals for modern devices
  • Inconsistent documentation leads to repair mistakes and wasted parts
  • High training costs prevent new technicians from entering the field
  • Language barriers with English-only manufacturer guides
  • 60% of e-waste in Nigeria comes from devices deemed "unrepairable"

💡 Our Solution

Artifex democratizes repair expertise by turning any damage photo into instant, professional repair guides powered by Google Gemini 2.5 Pro.

How It Works

  1. Authenticate – Secure Auth0 login for authorized users
  2. Upload Photo – Snap or upload device damage image (up to 4MB)
  3. Get Expert Guide – Receive step-by-step instructions with visual references in 30 seconds

✨ Key Features

🤖 Google Gemini 2.5 Pro Integration

  • Multimodal vision analysis detects device type, damage severity, and repair complexity
  • Structured JSON output generates consistent repair documentation
  • Safety-first approach provides hazard warnings specific to detected damage
  • Context-aware instructions adapt to damage patterns and device types

📸 Intelligent Visual References

  • Pexels API integration fetches contextual repair images
  • Shows relevant tools, components, and techniques based on device type
  • High-quality imagery improves comprehension for visual learners

🔐 Enterprise-Grade Security

  • Auth0 authentication protects proprietary repair knowledge
  • Session management for repair shop team workflows
  • Secure API routes prevent unauthorized access

📋 Comprehensive Repair Guides

Each guide includes:

  • ⚠️ Safety warnings and hazard callouts
  • 🛠️ Required tools and materials lists
  • 📸 Contextual repair images
  • 📝 Step-by-step instructions
  • ⏱️ Time estimates and difficulty ratings

🎨 Field-Optimized Design

  • Mobile-first responsive works on any device
  • Auto-scroll navigation keeps focus on current step
  • High contrast interface for outdoor visibility
  • Minimal text design for quick scanning

🚀 Tech Stack

Frontend

  • Next.js 16 – App Router with Turbopack for optimal performance
  • Tailwind CSS v4 – Custom inline theme for modern styling
  • lucide-react – Clean, consistent iconography

Backend & AI

  • Google Generative AI SDK – Gemini 2.5 Pro with vision capabilities
  • Pexels API – Contextual repair image search
  • Vercel Serverless Functions – Scalable API routes

Security & Performance

  • Auth0 SPA SDK – Production-ready authentication
  • Vercel Deployment – Global edge network for fast load times
  • Vercel Speed Insights – Real-time performance monitoring

🎥 Demo Video

Watch Artifex in action: DEMO

Key Moments:

  • 0:00 – Problem overview
  • 0:30 – Live damage photo upload
  • 1:00 – AI-generated repair guide
  • 1:45 – Technical architecture
  • 2:30 – Community impact

🏆 Why Artifex Wins

✅ Gemini API Mastery

  • Advanced multimodal capabilities (image + text analysis)
  • Structured output parsing for reliable documentation
  • Real-world application of vision AI

✅ Community Impact

  • Empowers 500+ GSU Computer Science students
  • Supports 20+ local repair shops in Gombe
  • Reduces e-waste through accessible repair knowledge

✅ Technical Excellence

  • Production-ready with secure authentication
  • Performance-optimized for mobile networks
  • Clean, maintainable codebase

✅ Innovation

  • First AI repair assistant designed for African field technicians
  • Novel combination: Vision AI + contextual image search + safety analysis
  • Solves real problem with scalable solution

📊 Impact Metrics

  • ⚡ 30-second turnaround vs 2+ hours manual documentation
  • 🎓 500+ students gain repair resources
  • 🔧 20+ repair shops can standardize workflows
  • 🌍 Reduces e-waste by making repairs accessible

🛠️ Local Setup

# Clone repository
git clone https://github.com/saviour2/Artifex
cd artifex-repair-assistant

# Install dependencies
npm install

# Configure environment variables
cp .env.example .env.local
# Add Auth0, Gemini, and Pexels credentials

# Run development server
npm run dev

Required Environment Variables

# Auth0
NEXT_PUBLIC_AUTH0_DOMAIN=your-domain.auth0.com
NEXT_PUBLIC_AUTH0_CLIENT_ID=your-client-id
NEXT_PUBLIC_AUTH0_REDIRECT_URI=http://localhost:3000

# Google Gemini
GEMINI_API_KEY=your-gemini-api-key

# Pexels
PEXELS_API_KEY=your-pexels-api-key

👥 Team

Saikat Das – Full-Stack Development & AI Integration
Subarna Maity – UI/UX Design & Full-Stack Soumyadeep Dey – Backend Architecture & API Integration

🔗 Links

Built With

Share this project:

Updates