Automated Real Estate Sourcing Tool Development
Budget: $1,500 – $3,000 USD
Description:
We are seeking a skilled developer to build the first version of a real estate deal-sourcing tool. This project is Phase 1–3 of a larger roadmap.
The tool will automatically collect property data from multiple sources across the U.S., normalize it into a database, enrich it with AI, apply scoring criteria, and generate daily ranked reports.
Project Scope (Phase 1–3):
• Data Ingestion (APIs + Scraping):
• Integrate with available APIs (county assessor & tax record APIs, Census Bureau API, SEC/EDGAR filings).
• Build scrapers for sources without APIs (LoopNet, Crexi, Zillow Commercial, etc.).
• Database:
• Design schema to store properties, sources, listings, and historical records.
• Implement deduplication using address/geolocation matching.
• LLM Enrichment:
• Use ChatGPT/OpenAI (or other LLMs) to parse unstructured text (e.g., “leases expire in 2027” → structured field).
• Scoring Engine:
• Apply configurable weighted criteria (e.g., lease expirations, occupancy rates, rent gap, distress signals).
• Reporting:
• Generate automated daily reports (CSV + PDF/HTML).
• Reports should be ranked by score.
• Deliver reports via email or dashboard export.
• Automation:
• Set up scheduler for daily refreshes.
• Include logging + error tracking.
Deliverables:
• API + scraper hybrid pipeline.
• Structured database with historical tracking.
• LLM integration for text parsing.
• Scoring engine with adjustable weights.
• Automated daily ranked reports (CSV + PDF).
• Documentation for setup and extension.
Required Skills:
• Python (APIs, Scraping: Requests, Scrapy, BeautifulSoup, Selenium).
• REST/GraphQL API integration.
• Database design (Postgres or SQLite).
• LLM integration (OpenAI API, LangChain, or similar).
• Data cleaning + deduplication.
Nice-to-Have Skills:
• GIS/Geocoding (address matching).
• Familiarity with U.S. real estate datasets.
• Experience with Airflow or other schedulers.
We are seeking a skilled developer to build the first version of a real estate deal-sourcing tool. This project is Phase 1–3 of a larger roadmap.
The tool will automatically collect property data from multiple sources across the U.S., normalize it into a database, enrich it with AI, apply scoring criteria, and generate daily ranked reports.
Project Scope (Phase 1–3):
• Data Ingestion (APIs + Scraping):
• Integrate with available APIs (county assessor & tax record APIs, Census Bureau API, SEC/EDGAR filings).
• Build scrapers for sources without APIs (LoopNet, Crexi, Zillow Commercial, etc.).
• Database:
• Design schema to store properties, sources, listings, and historical records.
• Implement deduplication using address/geolocation matching.
• LLM Enrichment:
• Use ChatGPT/OpenAI (or other LLMs) to parse unstructured text (e.g., “leases expire in 2027” → structured field).
• Scoring Engine:
• Apply configurable weighted criteria (e.g., lease expirations, occupancy rates, rent gap, distress signals).
• Reporting:
• Generate automated daily reports (CSV + PDF/HTML).
• Reports should be ranked by score.
• Deliver reports via email or dashboard export.
• Automation:
• Set up scheduler for daily refreshes.
• Include logging + error tracking.
Deliverables:
• API + scraper hybrid pipeline.
• Structured database with historical tracking.
• LLM integration for text parsing.
• Scoring engine with adjustable weights.
• Automated daily ranked reports (CSV + PDF).
• Documentation for setup and extension.
Required Skills:
• Python (APIs, Scraping: Requests, Scrapy, BeautifulSoup, Selenium).
• REST/GraphQL API integration.
• Database design (Postgres or SQLite).
• LLM integration (OpenAI API, LangChain, or similar).
• Data cleaning + deduplication.
Nice-to-Have Skills:
• GIS/Geocoding (address matching).
• Familiarity with U.S. real estate datasets.
• Experience with Airflow or other schedulers.