Scale Real Estate Lead SaaS

I already have a working Python scraper, but it needs a full-scale makeover into a cloud-hosted SaaS that real estate professionals can rely on every day. Your mission is to keep what works, refactor what doesn’t, and wrap the whole engine in a polished web app.

Core upgrades I’m after
• Sharpen data-extraction accuracy so property details, owner contacts, and live market-trend metrics arrive clean and complete when we crawl major listing sites, property-management portals, and even MLS feeds.
• Boost processing speed. Large city-wide pulls should finish quickly without timing out or starving resources; asynchronous queues or lightweight microservices are welcome if they help.
• Replace the current bare-bones UI with an intuitive dashboard where users can filter, tag, and export leads in CSV/Excel or push directly to their CRM.

Tech freedom & expectations
The existing code is pure Python (BeautifulSoup, requests, basic pandas). I’m open to keeping that stack or shifting to Scrapy, Playwright, or headless browsers for tougher pages, as long as it remains maintainable. A modern backend (FastAPI, Django, or Flask) tied to a relational database is fine; React or Vue on the front end would be great for the dashboard. Containerisation (Docker) and a CI/CD pipeline are a plus because I’ll eventually host on AWS.

Deliverables
1. Refactored scraper engine with configurable source modules for listing sites, property-management sites, and MLS data.
2. Scalable processing layer that queues jobs and writes structured results to the database.
3. Web dashboard: search, filter, tag, and export leads.
4. Admin panel with user management, credit/usage tracking, and simple role controls.
5. Setup scripts, environment files, and concise developer documentation so future iterations are straightforward.

Acceptance
If I can spin up the stack via a single command, run a crawl on three demo sources, and export a clean CSV in under ten minutes, we’re good.