Sekirkallc/ai-data-factory-real-estate

AI Data Factory — Real Estate Dataset

Autonomous AI Data Factory for RAG and AI Agents

High-quality synthetic real estate property dataset automatically generated and updated hourly via GitHub Actions, published to Hugging Face for AI training, retrieval-augmented generation (RAG), and agent training.

📊 Dataset Overview

• —Total Records: Continuously growing (100+)
• —Update Frequency: Hourly (automated via GitHub Actions)
• —License: MIT (Commercial use allowed)
• —Format: JSONL and Parquet
• —Source: https://github.com/sekirkallc/ai-data-factory

📁 Files

• —latest.jsonl — All records in JSONL format (recommended for loading)
• —latest.parquet — All records in Parquet format (columnar, efficient)
• —data_YYYYMMDD_HHMMSS.jsonl — Timestamped snapshots
• —data_YYYYMMDD_HHMMSS.parquet — Timestamped snapshots

🔍 Schema

💡 Use Cases

• —Retrieval-Augmented Generation (RAG): Embeddings and semantic search for property queries
• —AI Agent Training: Real estate agent training and property matching
• —Fine-tuning: LLM fine-tuning on property data
• —Benchmarking: Testing retrieval and ranking systems
• —Synthetic Data: Foundation for augmentation and augmented datasets

🚀 Quick Start

Load with Pandas

import pandas as pd

# Load JSONL
df = pd.read_json('latest.jsonl', lines=True)
print(df.head())

# Load Parquet
df = pd.read_parquet('latest.parquet')
print(f"Total records: {len(df)}")