EstateForge AI
EstateForge automates property search and analysis in four stages: 1. **Search** — Firecrawl extracts listings from selected real estate websites 2. **Extract** — structured property data via Pydantic schemas (address, price, beds, baths, sqft) 3. **Analyze** — Market Analysis Agent provides neighborhood and investment insights 4. **Value** — Property Valuation Agent assesses each listing for investment potential
EstateForge AI
Real estate search agent team by Sharad Khare.
EstateForge AI searches Zillow, Realtor.com, Trulia, and Homes.com using Firecrawl extraction, then delivers Gemini-powered market analysis and property valuations through a Streamlit interface.
Built and maintained by Sharad Khare — AI strategist, full-stack developer, and creator of practical agent workflows.
What this project does
EstateForge automates property search and analysis in four stages:
1. Search — Firecrawl extracts listings from selected real estate websites
2. Extract — structured property data via Pydantic schemas (address, price, beds, baths, sqft)
3. Analyze — Market Analysis Agent provides neighborhood and investment insights
4. Value — Property Valuation Agent assesses each listing for investment potential
Why Sharad Khare built EstateForge AI
Property hunting across multiple listing sites is fragmented and time-consuming. EstateForge demonstrates how Firecrawl extraction combined with Gemini agents can unify search, comparison, and investment analysis in one workflow.
How it works
City + State + Criteria + Website Selection
│
▼
Firecrawl Extraction (Zillow, Realtor, Trulia, Homes.com)
└── Structured property listings (Pydantic)
│
▼
Market Analysis Agent (Gemini 2.5 Flash)
└── Market trends and neighborhood insights
│
▼
Property Valuation Agent (Gemini 2.5 Flash)
└── Per-property investment assessments
│
▼
Streamlit tabs: Properties, Market Analysis, Valuations
Supported websites
| Website | Data Extracted |
|---------|---------------|
| Zillow | Listings, prices, beds/baths, sqft |
| Realtor.com | Listings, prices, property details |
| Trulia | Listings, neighborhood data |
| Homes.com | Listings, property features |
Features
- Multi-site property search with Firecrawl structured extraction
- Pydantic schemas for consistent property data
- Gemini 2.5 Flash agents for market analysis and valuations
- Streamlit UI with property comparison tabs and metrics
- Configurable search criteria: budget, type, beds, baths, sqft, features
Quick start
cd estateforge-ai
pip install -r requirements.txt
streamlit run app.py
Required API keys
- Google AI API key — Gemini agents for analysis and valuation
- Firecrawl API key — property listing extraction from real estate sites
Project structure
estateforge-ai/
├── app.py
├── estateforge/
│ ├── config.py
│ ├── schemas.py
│ └── search.py
├── requirements.txt
├── pyproject.toml
└── README.md
Who this is for
- Home buyers comparing properties across listing sites
- Real estate investors evaluating market opportunities
- Agents researching listings and neighborhood trends
- Developers learning Firecrawl + Gemini agent patterns for real estate
Use cases
- Search properties by city, state, budget, and criteria across multiple sites
- Compare listings side-by-side with structured data
- Get AI market analysis for target neighborhoods
- Receive per-property investment valuations and recommendations
Author
- Website: sharadkhare.in
License
MIT © Sharad Khare
Loading file tree…
Select a file to preview its source.
cd estateforge-ai
pip install -r requirements.txt
streamlit run app.py