4stene
In development · not yet open

Real estate · Slovenia

The real-estate platform that already knows the property.

4stene is a CRM for Slovenian real-estate agencies, a listings marketplace, and the national land registry underneath both. Type an address and a listing starts with its exact location, parcels, size, floor and year built already filled in.

Get in touch We're building in private and will open to agencies later.

What we're building

A CRM for agents

Customers, buyer briefs, mandates and a deal pipeline from first viewing to signed agreement, shared across the agency with roles and permissions.

A listings marketplace

Listings that agencies and private sellers publish once, with photos processed automatically and every fact traceable to the registry.

Cadastre data underneath

Addresses, parcels, buildings and recorded sale prices from GURS, Slovenia's surveying and mapping authority, synced nightly and drawn on the map.

On the roadmap Estimated property prices Publishing to other portals Floor plan scans 3D walkthroughs from photos Agency websites AI listing translation

Check, don't retype

Most of what a listing needs is already on record. Agents start from the registry and only add what it can't know.

  1. 01
    Type the address
    Autocomplete finds the building from the national address register.
  2. 02
    Pick the unit
    Size, floor, year built, elevator, parcels and past sale prices fill in on their own.
  3. 03
    Add photos and publish
    Correct anything that changed, add photos from the phone, and the listing goes live.

Who's behind it

Portrait of Primož Lavrič
Primož Lavrič
Founder · Ljubljana

Software architect with a decade of experience across real-time control systems, scientific instrumentation, medical devices and cloud-native backends. He has owned full product architecture from day one, led teams of up to six engineers, and shipped production systems in Go, C++ and TypeScript. He designs and builds 4stene end to end, from the GURS data pipeline to the agents' dashboard.

Go C++ TypeScript Python AWS DDD and CQRS Real-time systems Machine learning
MSc and BSc in Computer and Information Science, University of Ljubljana

Experience

  • Staff software engineer · Teads2026–present
    Building the integration of connected TV (CTV) advertising data with Google Analytics 4.
  • Senior software engineer · Veza (ServiceNow)2026
    Backend features in Go across the Access Reviews product, on Kubernetes with OpenTelemetry.
  • Founding engineer and tech lead · Valhalla Turrets2024–2025
    Built the software department from scratch to six engineers and designed the core real-time control system alone; shipped a prototype in year one and the first product in year two.
  • Staff software engineer · The NU2022–2024
    Owned three Go microservices on AWS (Lambda, SQS, SNS, S3, SageMaker), two of them core to the product, and led a team of four.
  • ML engineer · Soniox2022
    Training-data pipelines for a state-of-the-art speech recognition model.
  • Senior engineer and architect · Cosylab2017–2022
    Architected control software for ITER, IPP Greifswald and MedAustron, and led a PET scanner data-acquisition project for Positrigo.