Project Finder
Real Estate Agency Platform
A subscription-based real estate platform that allows users to discover and browse properties. Users receive notifications from admins, use advanced filters, and access premium listings after subscribing.
What was the measurable impact?
Thousands of property views per month with real-time updates
What problem did Project Finder solve?
Users struggled to find properties effectively, and agencies had no streamlined method to deliver targeted property alerts.
How was Project Finder solved?
Developed a web and mobile-friendly platform with subscription access, advanced property filtering, and admin push notifications to highlight listings.
How is Project Finder architected?
Dedicated Search Engine Offloading (Meilisearch) + Asynchronous Notification Queues
Decoupled search execution from primary MongoDB datastore to deliver instant sub-15ms multi-criteria property search results.
- Meilisearch engine integration for fuzzy text search & faceted filtering
- BullMQ background worker queue for bulk push notification broadcasting
- Tiered subscription access middleware enforcing role-based premium listing visibility
What was hard to build in Project Finder?
Challenge 1
Heavy database CPU load when users executed complex range and keyword filters on property listings.
How it was solved
Offloaded all property search indexing to Meilisearch with MongoDB change streams auto-syncing real-time property updates in the background.
Challenge 2
Synchronous push notification dispatches blocking HTTP response times when admins posted hot deal alerts.
How it was solved
Offloaded dispatch tasks to background queues powered by Redis and BullMQ, allowing instant HTTP acknowledgement while worker processes handled bulk notification pushes.
What can Project Finder do?
- Subscription-based property access
- Advanced property filtering (price, location, type)
- Admin push notifications
- Property search and saved listings
- User reviews and ratings
- Secure account management and authentication