Dallal.

Dallal is a PropTech platform built at Metafic: geospatial property search, developer management, property exchange, and the approval workflows that tie them together.

I lead a team of developers delivering Dallal on Node.js and PostgreSQL, and I architected its core modules — the geospatial search, developer management, property exchange and approvals.

Buyers search the way they think about property: pan and zoom a map, or draw their own search zone. That needs real spatial queries, so the map runs on PostgreSQL with PostGIS — GiST indexes for bounding-box searches as the viewport moves, KNN queries for "nearest first", and polygon queries for hand-drawn zones.

The database does the geometry. Nothing fetches every listing to measure distances in application code, so search stays fast as listings grow.

Every step in an exchange or approval has to reach the right person. I built a multilingual notification and communication engine — email and push notifications covering 20+ transactional workflows, processed as background jobs so API responses never wait on delivery.

The platform integrates Firebase, OpenAI, payment gateways and other external APIs. A steady part of the work is database performance, security and keeping high-traffic API responses fast.

Search, listings, exchanges and approvals run through one set of workflows on a spatial data model built for map-first search, and the team ships new modules on top of it.