🦜 Sendbird AI Agent (RAG & Tool Calling)

Sendbird Chat API와 LangChain을 활용하여 구축한 지능형 CS 에이전트입니다. 단순한 규칙 기반 챗봇을 넘어, 기업 문서를 참조(RAG)하여 답변하고 실제 비즈니스 로직(Tool Calling)을 수행할 수 있도록 설계되었습니다.

🚀 Key Features

• Real-time Communication: Sendbird Webhook & API를 활용한 실시간 양방향 통신
• RAG (Retrieval-Augmented Generation): LangChain과 Vector DB(Chroma)를 활용하여 PDF/TXT 문서 기반의 정확한 답변 제공
• Prompt Engineering: System Prompt를 활용한 페르소나 부여 및 Hallucination 제어
• Async Processing: FastAPI의 BackgroundTasks를 활용한 비동기 메시지 처리 및 성능 최적화
• Scalable Architecture: 기능 확장(Tool Calling)이 용이한 모듈형 구조

🛠 Tech Stack

• Framework: Python 3.11, FastAPI
• AI & LLM: LangChain, OpenAI (GPT-3.5/4), ChromaDB (Vector Store)
• Infrastructure: Ngrok (Tunneling), Docker (Planned)
• Communication: Sendbird Chat SDK / API

🏗 Architecture

graph LR
    User[User] -->|"Send Message"| Sendbird["Sendbird Platform"]
    Sendbird -->|"Webhook (POST)"| Agent["FastAPI Agent Server"]
    Agent -->|"Retrieve Context"| VectorDB[("Chroma DB")]
    VectorDB -->|"Context"| Agent
    Agent -->|"Prompt + Context"| LLM["OpenAI GPT"]
    LLM -->|"Answer"| Agent
    Agent -->|"Send API"| Sendbird
    Sendbird -->|"Reply"| User

Loading

⚡️ Quick Start

1. Prerequisites

• Python 3.9+
• Sendbird Application ID & API Token
• OpenAI API Key

2. Installation

# Clone the repository
git clone https://github.com/YOUR_USERNAME/sendbird-ai-agent.git
cd sendbird-ai-agent

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

3. Environment Setup

Copy .env.example to .env and fill in your credentials:

cp .env.example .env

Edit .env file:

SENDBIRD_APP_ID=your_sendbird_app_id
SENDBIRD_API_TOKEN=your_sendbird_api_token
OPENAI_API_KEY=sk-your_openai_api_key

4. Run Backend Server

# Start FastAPI server
uvicorn main:app --reload --port 8001

# Or with custom host/port
uvicorn main:app --host 0.0.0.0 --port 8001

The API will be available at http://localhost:8001

5. Run Dashboard (Optional)

cd dashboard
npm install
npm run dev

The dashboard will be available at http://localhost:3000

📚 API Endpoints

• GET / - Health check
• POST /webhook - Sendbird webhook handler
• GET /api/logs - Get all conversation logs
• PUT /api/logs/{log_id}/feedback - Update feedback for a log

🧪 Running Tests

pytest tests/

📝 License

This project is licensed under the MIT License.