Freelance Project
Oslo Propiedades: Real Estate Digital Ecosystem
Three projects for a real estate agency: a landing built from scratch with Astro, a WordPress to Next.js migration with CRM integration, and a scraper that feeds an AI chat.
Tech stack
The technologies used and why.
The context
Oslo Propiedades is a real estate agency that needed to modernize its digital presence. The starting point was a slow WordPress site, with no CRM integration and an outdated landing page. The work was organized into three projects developed in parallel.
Landing with Astro
The first project was a brand new landing page from scratch. The design agency delivered the mockups in Photoshop and my job was to translate them into code: pixel-perfect, responsive and fast-loading. I chose Astro for the advantages of static generation: no unnecessary JavaScript, minimal load times and easy to maintain.
Migration to Next.js with Tokko CRM
The most complex project. Oslo uses Tokko Broker as its real estate CRM: that is where all the properties, their photos, prices and availability status live. The new Oslo website consumes the Tokko API as its single source of truth, replacing the manual data entry that WordPress required.
The architecture uses Next.js with App Router, Zustand for the global state of search filters, Leaflet for the interactive property map, and Zod to validate the API responses before rendering them.
Scraping and data for the AI chat
The third project solves a synchronization problem: not every property on the current website is in Tokko, and vice versa. I built a scraping script that crawls the production website, extracts all the properties and developments, and compares them against the Tokko data. The result is a consolidated and clean JSON.
That JSON feeds an AI chat integrated into the Oslo website. The chat can answer questions about available properties, filter by area or price and provide contact information, using the real and up-to-date catalog as its knowledge base.