🇫🇷 Résumé (FR) : Application web intégrant l'intelligence artificielle générative pour automatiser et optimiser les réponses (support client, prospection, ou requêtes métier). Le projet démontre une capacité à connecter une interface moderne (React/Next.js) à des modèles de langage (LLMs) pour résoudre un problème de productivité concret.
An AI-driven application designed to scale customer interactions and automate responses using advanced LLM pipelines.
🔗 Live Application: replai-two.vercel.app
Handling incoming queries (whether for customer support, lead generation, or internal requests) at scale is incredibly time-consuming. Businesses either rely on rigid, frustrating decision-tree chatbots or expensive manual labor, leading to slow response times and dropped leads.
Replai acts as an intelligent intermediary. It ingests prompts or incoming messages and leverages Generative AI to craft context-aware, highly relevant responses in seconds. It bridges the gap between complex AI capabilities and a simple, user-friendly interface.
- Contextual Generation: Uses system prompts to ensure the AI's tone and factual output align with business needs.
- Seamless UI/UX: A clean, responsive frontend built for speed, allowing users to generate and copy/export replies instantly.
- Production-Ready: Fully deployed and optimized for edge performance.
| Component | Technology | Purpose |
|---|---|---|
| Frontend | React / Next.js | Dynamic, responsive user interface. |
| Styling | Tailwind CSS | Modern, utility-first styling for rapid UI development. |
| AI Integration | LLM APIs (Gemini / OpenAI) | Natural language understanding and text generation. |
| Deployment | Vercel | CI/CD pipeline and fast edge-network hosting. |
- API Route Handling: Securely managing API keys and server-side requests to prevent exposing LLM credentials on the client side.
- State Management: Handling asynchronous API calls, loading states, and error catching gracefully within the React lifecycle to ensure a smooth user experience.
- (Optional: If you integrated Make.com or Supabase for this project too, mention it here: "Data Persistence with Supabase" or "Webhook triggers via Make.com").
# Clone the repository
git clone [https://github.com/juniorbaw/replai.git](https://github.com/juniorbaw/replai.git)
# Navigate to the project directory
cd replai
# Install dependencies
npm install
# Set up environment variables
# Create a .env.local file and add your AI API keys (e.g., NEXT_PUBLIC_API_KEY)
# Start the development server
npm run dev