title	DubaiNest AI
emoji	🌇
colorFrom	yellow
colorTo	blue
sdk	docker
pinned	true
license	mit
short_description	Dubai Real Estate Assistant
thumbnail	https://cdn-uploads.huggingface.co/production/uploads/681ecc9f68e6f14b1e7d43b0/NxTSoa93VtJBp5NvJUb_I.png

🌇 DubaiNest AI — Dubai Real Estate Assistant

An end-to-end RAG-powered AI chatbot that answers questions about Dubai rental prices, area comparisons, RERA tenant rights, and move-in costs — built as part of the AI Accelerator Bootcamp by Decoding Data Science.

---

🎯 What it does

DubaiNest AI is a production-ready AI assistant that helps property seekers, expats, and investors in Dubai navigate one of the world's most complex real estate markets — through a simple chat interface.

Ask it things like:

• "What is the average rent for a 1BR in JVC?"
• "Can my landlord increase rent by 20%?"
• "What is the total move-in cost for a AED 90,000 flat?"
• "Compare JVC vs Dubai Marina for a young professional"
• "Which areas have metro access under AED 80,000/year?"

---

🏗️ Architecture

User Question
      ↓
Flask API (Waitress WSGI)
      ↓
LlamaIndex Query Engine
      ↓
OpenAI Embeddings (text-embedding-3-small)
      ↓
Pinecone Vector Store — semantic search → top 4 chunks
      ↓
Custom Prompt Template + GPT-4o-mini
      ↓
Answer → Chat UI (served from Flask)

---

🛠️ Tech Stack

Core AI & RAG

Backend & Deployment

Frontend

Also Used

---

📁 Project Structure

DubaiNestAI/
├── app_llama.py          # LlamaIndex RAG pipeline (active)
├── app.py                # LangChain RAG pipeline (backup)
├── Dockerfile            # Docker build config
├── requirements.txt      # Python dependencies
├── README.md
├── .gitignore
├── static/
│   └── index.html        # Chat UI
└── data/
    ├── area_guide.csv    # Dubai area rental data (20+ areas)
    ├── rera_rules.txt    # RERA tenant law rules
    └── cost_logic.txt    # Move-in cost calculations

---

📊 Knowledge Base

The AI answers are grounded in a structured knowledge base covering:

• 20+ Dubai areas — JVC, Dubai Marina, Downtown, Business Bay, Palm Jumeirah, Dubai Hills, and more
• Rental prices — Studio, 1BR, 2BR, 3BR, Villa by area (2025 data)
• RERA regulations — Rent increase rules, tenant rights, eviction laws, dispute resolution
• Move-in cost logic — Security deposit, agency fee, DEWA, chiller, total calculations

---

🚀 Run Locally

1. Clone the repo

git clone https://github.com/NipunKavinda95/DubaiNestAI.git
cd DubaiNestAI

2. Create .env file

OPENAI_API_KEY=sk-...
PINECONE_API_KEY=pcsk_...
PINECONE_INDEX=...

3. Install dependencies

pip install -r requirements.txt

4. Run the app

python app.py

5. Open browser

http://localhost:7860

---

🔄 LangChain → LlamaIndex Migration

This project was intentionally built with both frameworks to demonstrate the migration path:

	LangChain (app_lang.py)	LlamaIndex (app.py)
Loader	CSVLoader, TextLoader	SimpleDirectoryReader
Splitter	RecursiveCharacterTextSplitter	SentenceSplitter
Vector Store	ChromaDB (local)	Pinecone (cloud)
Query	rag_chain.invoke()	query_engine.query()
Prompt vars	{context} {question}	{context_str} {query_str}

Switch between them in Dockerfile:

CMD ["python", "app.py"]   # LlamaIndex (active)
CMD ["python", "app_lang.py"]         # LangChain (backup)

---

🌐 Deployment

Deployed on HuggingFace Spaces using Docker SDK.

	Detail
Live URL	https://huggingface.co/spaces/nipunkavindaAI/DubaiNestAI
Port	7860 (HuggingFace default)
Server	Waitress (production WSGI)
Secrets	OPENAI_API_KEY, PINECONE_API_KEY via HF Space Secrets

---

👨‍💻 Built By

Nipun Kavinda I am Nipun Kavinda, a Mechanical Automation & Maintenance Engineer now specialising in Industrial AI and intelligent automation. My engineering foundation gives me something most software developers do not have — a deep understanding of real-world physical systems, machinery, and control logic. Dubai, UAE 🇦🇪