Real estate
Property Investment Analytics Platform
Combines auction listings, rental comparables and bank loan packages to show which properties would pay for themselves.
The problem
Auction listings, rental prices and loan eligibility live in three different places. Investors and agents wanted one answer for every property: would the rent cover the monthly instalment, and by how much?
What I built
- Scheduled data pipelines that collect property listings and housing-loan packages, then reconcile auction records with rental comparables.
- A monthly profit figure for every property, based on the cheapest loan it qualifies for rather than a flat assumed rate.
- A shareable public portal that shows only profitable matches, with built-in advertising slots.
- An internal workspace for agents: project listings, contacts, contracts, advertisement generation, dashboards, and self-service SQL and AI analysis.
- Users, roles, permissions and teams, integrated with Google Workspace.
Technical highlights
- Separate fast and slow Celery queues keep long browser-automation jobs from delaying quick tasks.
- Ad-hoc SQL analysis runs through a read-only database role.