A Next.js-powered intelligent support chat application featuring a Retrieval-Augmented Generation (RAG) system powered by Google's Gemini LLM for context-aware, knowledge-based responses.

RAG Support Chat is an advanced AI-powered customer support solution that combines the power of Large Language Models (LLM) with Retrieval-Augmented Generation (RAG) technology. Unlike traditional chatbots, this system retrieves relevant information from your custom knowledgebase and uses it to generate accurate, contextual responses through semantic search and AI reasoning.

• 🧠 Retrieval-Augmented Generation: Combines semantic search with LLM reasoning for accurate responses
• 🔍 Semantic Search: Advanced document retrieval using embedding-based similarity matching
• 📚 Custom Knowledgebase Integration: Upload and index your own documentation for domain-specific responses
• 🎯 Context-Aware Responses: AI generates answers based on retrieved relevant documents
• 🔄 Dynamic Knowledge Retrieval: Real-time document search for each user query

• 💬 Intelligent Chat Interface: Modern, responsive chat UI with real-time messaging
• ⚙️ Easy Configuration: Simple setup for API keys and knowledgebase management
• 🎨 Modern Design: Built with Tailwind CSS and shadcn/ui components
• 📱 Responsive Layout: Works seamlessly across desktop and mobile devices
• 🌙 Dark Mode Support: Elegant theme switching capabilities

1. Document Ingestion: Your knowledgebase content is processed and indexed
2. Query Processing: User questions are analyzed for semantic meaning
3. Information Retrieval: Relevant documents are retrieved using semantic search
4. Context Augmentation: Retrieved content is formatted as context for the LLM
5. Response Generation: Gemini LLM generates accurate responses based on the context
6. Answer Delivery: Contextual, knowledgebase-backed responses are delivered to users

• Customer Support: Automated, accurate responses to customer inquiries
• Internal Knowledge Base: Employee self-service for company information
• Product Documentation: Interactive documentation assistance
• FAQ Automation: Dynamic FAQ responses based on your content
• Training and Onboarding: Interactive learning assistance

• Accuracy: Responses are grounded in your actual documentation
• Relevance: Semantic search ensures contextually appropriate information retrieval
• Scalability: Easy to update knowledgebase without retraining models
• Cost-Effective: Leverages existing LLMs while maintaining domain expertise
• Transparency: Clear connection between responses and source documents

• Frontend: Next.js 14+ with TypeScript and Tailwind CSS
• LLM Integration: Google Gemini API for natural language processing
• Semantic Search: Custom implementation for document retrieval
• State Management: React Context for configuration and chat state
• UI Components: shadcn/ui for consistent, accessible interface

• Node.js 18+
• npm, yarn, pnpm, or bun
• Google Gemini API key (for LLM capabilities)
• Your knowledgebase content (documents, FAQs, support articles, etc.)

1. Clone the repository:

git clone <your-repo-url>

2. Install dependencies:

npm install
# or
yarn install
# or
pnpm install
# or
bun install

3. Create a .env.local file in the root directory:

# Add any required environment variables here
NEXT_PUBLIC_APP_URL=http://localhost:3000

4. Run the development server:

npm run dev
# or
yarn dev
# or
pnpm dev
# or
bun dev

5. Open http://localhost:3000 with your browser to see the result.

• Retrieval-Augmented Generation Paper
• Google Gemini API Documentation
• Semantic Search Best Practices

• Next.js Documentation
• Tailwind CSS
• shadcn/ui