🏢 SuitSenseAI - AI-Powered Real Estate Intelligence Platform

Conversational AI for Real Estate Analytics & Market Intelligence

🌟 What is SuitSenseAI?

SuitSenseAI is a cutting-edge conversational AI platform that revolutionizes real estate analytics. Users can ask questions in plain English and receive sophisticated insights, interactive visualizations, and professional reports about condominium markets.

✨ Key Features

🤖 Conversational Analytics - Ask questions in natural language, get intelligent answers
📊 Dynamic Visualizations - Auto-generated charts, graphs, and interactive maps
📋 AI-Generated Reports - Professional PDF reports created on-demand
🗺️ Location Intelligence - Google Maps integration with proximity analysis
📈 Market Analysis - Holding periods, sales volumes, and trend predictions
🔒 Secure Code Execution - Safe AI-generated Python code execution
💬 Memory-Aware - Maintains conversation context for complex analysis

SuitSenseAI_Demo - Watch Video

💡 Usage Examples

Queries

"Which buildings has the highest sale in collins"
"can u put this into HTML Table"
"Can u please generate a graph of this for me to analyze"
"can u please add a third row with avergae median sale"
"can u also show me the closest school distance to each of this building"
"can u show me on map the nearby schools around this property"
"can u also add a new column with average holding period"
"provide me a pdf report of the whole table to download"

🏗️ Architecture & Technology Stack

🧠 AI & Machine Learning

Google Gemini 2.0 Flash Experimental - Latest multimodal LLM
LangGraph ReAct Agent - Advanced reasoning and acting framework
FAISS Vector Database - Semantic search capabilities
Google Embeddings - High-quality text embeddings

🗄️ Data & Backend

PostgreSQL - Production-grade database with geographic data
LangChain SQL Toolkit - Intelligent database interactions
Flask Web Framework - Lightweight and scalable web server
Session Management - Conversation history and user tracking

🌐 APIs & Integrations

Google Maps JavaScript API - Interactive mapping
Google Places API - Location search and discovery
Google Directions API - Distance calculations
Google Geocoding API - Address to coordinate conversion

📊 Visualization & Reporting

Chart.js - Dynamic chart generation
ReportLab - PDF report creation
Google Maps Advanced Markers - Custom map visualizations
Markdown Processing - Rich text formatting

🚀 Quickstart Guide

1. 🔹 Clone the Repository

git clone https://github.com/sohailshk/suitsense-ai.git
cd suitsense-ai

2. 🐍 Create Virtual Environment

python -m venv venv
# Windows
venv\Scripts\activate
# macOS/Linux
source venv/bin/activate

3. 🗄️ Setup PostgreSQL Database

-- From PostgreSQL shell (psql)
CREATE DATABASE suitsense_ai;
CREATE USER readonly_user WITH PASSWORD 'your_secure_password';
GRANT CONNECT ON DATABASE suitsense_ai TO readonly_user;
GRANT USAGE ON SCHEMA public TO readonly_user;
GRANT SELECT ON ALL TABLES IN SCHEMA public TO readonly_user;
ALTER DEFAULT PRIVILEGES IN SCHEMA public GRANT SELECT ON TABLES TO readonly_user;

4. 🔍 Configure Environment Variables

Windows (PowerShell):

$env:GPLACES_API_KEY="your_google_places_api_key"
$env:GEMINI_API_KEY="your_gemini_api_key"
$env:FLASK_SECRET="your_flask_secret_key"
$env:PG_USER="readonly_user"
$env:PG_PASSWORD="your_secure_password"
$env:PG_PORT="5432"
$env:PG_DB="suitsense_ai"

macOS/Linux (Bash):

export GPLACES_API_KEY="your_google_places_api_key"
export GEMINI_API_KEY="your_gemini_api_key"
export FLASK_SECRET="your_flask_secret_key"
export PG_USER="readonly_user"
export PG_PASSWORD="your_secure_password"
export PG_PORT="5432"
export PG_DB="suitsense_ai"

5. 📃 Load Sample Data

psql -d suitsense_ai < sample_db.sql

6. ⚖️ Install Dependencies

pip install -r requirements.txt

7. 🧪 Test Your Setup

python test_gemini.py

You should see all green checkmarks ✅

8. 🚦 Run the Application

python server.py

Visit http://localhost:5000 and start asking questions! 🎉

🔑 API Keys Setup

Google APIs (Required)

Google Cloud Console: https://console.cloud.google.com/
Enable APIs:
- Google Places API
- Google Maps JavaScript API
- Google Geocoding API
- Google Directions API
Create API Key and restrict it to your domains

Google Gemini AI (Required)

Google AI Studio: https://aistudio.google.com/
Get API Key for Gemini 2.0
Set Usage Limits as needed

🏗️ Project Structure

SuitSenseAI/
├── 📄 main.py              # Core AI processing logic
├── 🌐 server.py            # Flask web application
├── 🛠️ tools.py             # AI tools and integrations
├── 📝 prefix.py            # AI system prompts and instructions
├── 🔧 boilerplate.py       # Code templates and snippets
├── 🧪 test_gemini.py       # Integration testing
├── 🗄️ init_db.py           # Database initialization
├── 📊 sample_db.sql        # Sample real estate data
├── 📋 requirements.txt     # Python dependencies
├── 📖 GEMINI_MIGRATION.md  # Migration guide from OpenAI
├── 📁 templates/           # HTML templates
│   └── 🏠 index.html       # Main web interface
├── 📁 static/              # Static assets
│   ├── 🖼️ images/          # Logo and images
│   └── 📄 README.md        # Static files documentation
└── 📁 __pycache__/         # Python cache files

🧠 How It Works

1. Natural Language Processing

User questions are processed by Google Gemini 2.0 Flash Experimental, which understands real estate terminology and context.

2. Intelligent Tool Selection

The AI agent autonomously selects appropriate tools:

SQL queries for database analysis
Google Maps for location services
Vector search for semantic matching
Code generation for reports

3. Dynamic Content Generation

Based on the analysis, the system generates:

Interactive charts using Chart.js
Google Maps with custom markers
Professional PDF reports
Rich markdown responses

4. Security & Safety

All AI-generated code is scanned for malicious patterns before execution, ensuring a secure environment.

🎯 Advanced Features

🔍 Semantic Search

Uses FAISS vector database with Google embeddings for intelligent property name and address matching.

📊 Complex Analytics

Holding Period Analysis: Calculate investment holding periods
Market Segmentation: Analyze by unit type, building, or location
Geographic Intelligence: Distance calculations and proximity analysis
Trend Prediction: Historical analysis and forecasting

🗺️ Interactive Mapping

Advanced Google Maps Integration: Custom markers and clustering
School Proximity Analysis: Find nearby educational institutions
Distance Calculations: Multi-point route optimization
Boundary Visualization: Market area definitions

📋 Report Generation

PDF Creation: Executive-level reports with charts and analysis
Custom Layouts: Professional formatting with ReportLab
Data Visualization: Embedded charts and maps in reports

🔧 Configuration Options

Database Settings

# Modify in main.py for custom database connections
POSTGRES_USER = os.getenv("PG_USER")
POSTGRES_PASSWORD = os.getenv("PG_PASSWORD")
POSTGRES_PORT = os.getenv("PG_PORT")
POSTGRES_DB = os.getenv("PG_DB")

AI Model Configuration

# Modify in main.py for different Gemini models
llm = ChatGoogleGenerativeAI(
    model="gemini-2.0-flash-exp",  # or "gemini-pro", "gemini-2.0-flash"
    temperature=0.1,  # Adjust for creativity vs. consistency
)

Conversation Memory

# Modify in server.py for conversation history length
MAX_CONTEXT_LENGTH = 3  # Number of previous exchanges to remember

🧪 Testing & Development

Run Tests

# Test database connection
python test_db.py

# Test Gemini integration
python test_gemini.py

# Test Google Places API
python sandbox.py

Development Mode

# Run with debug mode
export FLASK_ENV=development
python server.py

🚀 Deployment

Production Setup

Use environment variables for all sensitive configuration
Set up proper database user permissions
Configure CORS for your domain
Use HTTPS for secure communication
Set up monitoring and logging

Docker Deployment (Optional)

# Example Dockerfile structure
FROM python:3.11-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY . .
EXPOSE 5000
CMD ["gunicorn", "--bind", "0.0.0.0:5000", "server:app"]

🤝 Contributing

We welcome contributions! Please see our contributing guidelines:

Fork the repository
Create a feature branch: git checkout -b feature/amazing-feature
Commit your changes: git commit -m 'Add amazing feature'
Push to the branch: git push origin feature/amazing-feature
Open a Pull Request

Development Guidelines

Follow PEP 8 style guidelines
Add tests for new features
Update documentation
Ensure security best practices

📝 License

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

🙏 Acknowledgments

Google AI for Gemini 2.0 and embeddings
LangChain for the agent framework
Google Maps Platform for location services
PostgreSQL for robust data storage
Open Source Community for amazing tools and libraries

📞 Support & Contact

GitHub Issues: Report bugs or request features
Documentation: Read the full docs
Email: sohail@example.com

⭐ Star this repository if you find it helpful!

🔄 Fork it to contribute or customize for your needs

📢 Share with others who might benefit from AI-powered real estate analytics

Made with ❤️ for the Real Estate & AI Community

Name		Name	Last commit message	Last commit date
Latest commit History 27 Commits
.github/workflows		.github/workflows
__pycache__		__pycache__
static		static
templates		templates
.dockerignore		.dockerignore
Dockerfile		Dockerfile
GEMINI_MIGRATION.md		GEMINI_MIGRATION.md
Procfile		Procfile
README.md		README.md
boilerplate.py		boilerplate.py
docker-compose.yml		docker-compose.yml
examples.py		examples.py
init_db.py		init_db.py
main.py		main.py
prefix.py		prefix.py
requirements.txt		requirements.txt
sample_db.sql		sample_db.sql
sandbox.py		sandbox.py
server.py		server.py
setup_env.py		setup_env.py
test_db.py		test_db.py
test_gemini.py		test_gemini.py
tools.py		tools.py

sohailshk/SuitSenseAI

Folders and files

Latest commit

History

Repository files navigation