🚗 CarBot Pro - Advanced Multi-Agent Car Sales System

A sophisticated multi-agent AI system for intelligent car sales, demonstrating advanced agent coordination, real-time research, and professional sales workflows.

Originally developed by Eduardo Hilario for AI Agents Day demo.

---

🎯 Overview

CarBot Pro showcases the cutting-edge capabilities of multi-agent AI systems in a real-world sales scenario. The system coordinates three specialized AI agents to provide a complete, professional car buying experience with intelligent inventory management, real-time research, and advanced negotiation capabilities.

🤖 Meet the Agents

Agent	Model	Role	Specialization
🎯 Carlos	GPT-4o	Sales Expert	Customer interaction, sales process, negotiation
🔬 María	o4-mini	Research Specialist	Vehicle research, technical analysis, market data
🏢 Manager	GPT-4o	Business Coordinator	Inventory management, pricing, business policies

✨ Key Features

🔧 Advanced Capabilities

• 🧠 Intelligent Agent Coordination - Seamless communication between specialized agents
• 🔍 Smart Inventory Search - AI-powered vehicle matching with 40+ enriched vehicle database
• 🌐 Real-time Web Research - Live vehicle information via SerpAPI integration
• 📊 Dynamic Customer Profiling - Automatic extraction and management of customer preferences
• 💼 Professional Sales Process - Complete sales funnel from greeting to closing
• 📈 Real-time Analytics - Comprehensive metrics and conversation analytics
• 🔄 State Management - Advanced sales stage tracking and progression

📋 Sales Process Stages

1. Greeting - Initial rapport building
2. Discovery - Needs assessment and profiling
3. Presentation - Intelligent vehicle recommendations
4. Objection Handling - Professional concern resolution
5. Negotiation - Policy-based pricing and alternatives
6. Closing - Sale finalization with inventory updates
7. Follow-up - Post-sale relationship management

🚗 Rich Vehicle Database

• Comprehensive Data: Make, model, year, color, mileage, price
• Technical Specs: Engine, transmission, fuel efficiency, safety ratings
• Special Features: Technology packages, interior details, location
• Price Range: €25,000 - €320,000 (economy to luxury supercars)

---

🚀 Quick Start

Option 1: Automated Setup (Recommended)

git clone https://github.com/LIDR-academy/car-salesman.git
cd car-salesman
python quick_setup_advanced.py

Option 2: Manual Installation

Prerequisites

• Python 3.8 or higher
• OpenAI API key (required)
• SerpAPI key (optional, for web research)

Installation Steps

1. Clone the repository

git clone https://github.com/LIDR-academy/car-salesman.git
cd car-salesman

2. Create virtual environment

python -m venv .venv

# Activate (Linux/macOS)
source .venv/bin/activate

# Activate (Windows)
.venv\Scripts\activate

3. Install dependencies

pip install --upgrade pip
pip install -r requirements.txt

4. Configure environment variables

cp config.env .env
# Edit .env with your API keys

5. Set up API keys in .env

# Required: OpenAI API Key
OPENAI_API_KEY=sk-your_openai_key_here

# Optional: SerpAPI Key for web research
SERPAPI_API_KEY=your_serpapi_key_here

6. Run the application

streamlit run enhanced_app.py

7. Access the interface
   • Open your browser to http://localhost:8501
   • Configure API keys in the sidebar
   • Click "🚀 Initialize Advanced System"
   • Start chatting with Carlos!

---

📖 Usage Examples

Basic Interaction Flow

# Example conversation flow
customer_input = "I'm looking for a safe family car"

# System processes through multi-agent workflow:
# 1. Carlos updates customer profile
# 2. Carlos consults Manager for inventory options
# 3. Manager searches inventory and provides recommendations
# 4. Carlos presents options to customer
# 5. Customer asks for technical details
# 6. Carlos requests research from María
# 7. María provides detailed analysis
# 8. Carlos shares processed information with customer

Demo Script Suggestions

1. Initial Contact

"Hi, I'm looking for a car"

Expected: Carlos greets and builds rapport

2. Family Needs

"I need a bigger, safer car because we just had a baby"

Expected: Carlos updates customer profile, shows understanding

3. Specific Requirements

"I want a red sedan that's less than 2 years old"

Expected: Carlos consults Manager, searches inventory

4. Brand Preference

"I'm interested in BMW vehicles"

Expected: Carlos refines search, presents BMW options

5. Technical Inquiry

"What safety features does it have for babies?"

Expected: Carlos requests research from María

6. Specific Details

"What's the trunk space in the BMW X3?"

Expected: María provides technical specifications

7. Pricing

"What's the price of the black BMW X3?"

Expected: Carlos consults Manager for official pricing

8. Negotiation

"Can you offer any discount?"

Expected: Manager evaluates policies, Carlos negotiates

9. Purchase Decision

"I'll take it"

Expected: Carlos finalizes sale, updates inventory

---

🏗️ Architecture

System Components

graph TB
    Client[👤 Customer]
    
    subgraph "Multi-Agent System"
        Carlos[🎯 Carlos<br/>Sales Agent<br/>GPT-4o]
        Maria[🔬 María<br/>Research<br/>o4-mini]
        Manager[🏢 Manager<br/>Coordinator<br/>GPT-4o]
        
        subgraph "Carlos Tools"
            T1[ConsultManager]
            T2[ResearchVehicleInfo]
            T3[UpdateCustomerProfile]
            T4[UpdateSalesStage]
            T5[RespondToClient]
            T6[FinalizeSale]
            T7[UpdateNotes]
        end
    end
    
    subgraph "External Systems"
        Inventory[🚗 Inventory DB]
        SerpAPI[🌐 SerpAPI]
        KB[📚 Knowledge Base]
    end
    
    Client <--> Carlos
    Carlos <--> T1
    Carlos <--> T2
    T1 <--> Manager
    T2 <--> Maria
    Manager <--> Inventory
    Maria <--> SerpAPI
    Maria <--> KB

Loading

---

🛠️ Technical Stack

Core Technologies

• 🤖 AI Models: OpenAI GPT-4o, o4-mini
• 🔗 Agent Framework: LangChain
• 🖥️ Frontend: Streamlit
• 📊 Data Processing: Pandas, NumPy
• 📈 Visualization: Plotly
• 🔍 Web Search: SerpAPI
• 🐍 Language: Python 3.8+

Key Dependencies

langchain>=0.3.25
langchain-openai>=0.3.18
langchain-community>=0.3.24
streamlit>=1.45.1
pandas>=2.2.3
plotly>=5.17.0
openai>=1.82.0
python-dotenv>=1.1.0

---

📁 Project Structure

carbot-pro/
├── 📄 README.md                          # This file
├── 🚀 quick_setup_advanced.py            # Automated setup script
├── 🎯 advanced_multi_agent_system.py     # Core multi-agent system
├── 📦 enhanced_inventory_manager.py      # Inventory management
├── 🖥️ enhanced_app.py                    # Streamlit interface
├── 🧪 test_system.py                     # System tests
├── 📋 requirements.txt                   # Python dependencies
├── ⚙️ config.env                         # Environment template
├── 📊 data/                              # Vehicle database
└── 📝 carbot_system.log                  # System logs

---

🔧 Configuration

Environment Variables

Variable	Required	Description
OPENAI_API_KEY	✅ Yes	OpenAI API key for GPT-4o and o4-mini
SERPAPI_API_KEY	❌ Optional	SerpAPI key for web research (fallback to knowledge base)

Agent Configuration

Each agent can be customized with different parameters:

# Carlos (Sales Agent)
carlos_llm = ChatOpenAI(
    temperature=0.8,      # Creative for sales
    model_name="gpt-4o",  # Latest GPT-4o
    max_tokens=1000
)

# María (Research Agent)
maria_llm = ChatOpenAI(
    temperature=1,        # Factual for research
    model_name="o4-mini", # Efficient for analysis
    max_tokens=800
)

# Manager (Coordinator)
manager_llm = ChatOpenAI(
    temperature=0.4,      # Balanced for decisions
    model_name="gpt-4o",  # Strategic thinking
    max_tokens=600
)

---

📊 Analytics & Monitoring

Available Metrics

• Conversation Analytics: Interaction count, agent communications
• Sales Performance: Stage progression, conversion tracking
• Customer Profiling: Profile completeness, preference analysis
• Agent Efficiency: Response times, tool usage patterns

Logging System

The system provides comprehensive logging:

• Agent Actions: All agent decisions and tool usage
• Inter-Agent Communications: Message flow between agents
• Customer Interactions: Complete conversation history
• System Events: Inventory updates, errors, performance metrics

---

🧪 Testing

Run the test suite to verify system functionality:

python test_system.py

The test suite covers:

• ✅ Agent initialization
• ✅ Tool functionality
• ✅ Inventory operations
• ✅ Customer profile management
• ✅ Sales stage transitions

---

🤝 Contributing

We welcome contributions! Here's how you can help:

Areas for Contribution

• 🔧 New Agent Tools: Expand Carlos's capabilities
• 🌐 Additional Research Sources: Integrate more data providers
• 📊 Enhanced Analytics: Advanced metrics and visualizations
• 🎨 UI Improvements: Better Streamlit interface
• 🧪 Testing: Expand test coverage
• 📚 Documentation: Improve guides and examples

Development Setup

1. Fork the repository
2. Create a feature branch: git checkout -b feature/amazing-feature
3. Make your changes
4. Add tests for new functionality
5. Commit your changes: git commit -m 'Add amazing feature'
6. Push to the branch: git push origin feature/amazing-feature
7. Open a Pull Request

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

⭐ If you find this project useful, please consider giving it a star on GitHub!

Built with ❤️ for the AI community