Our technology

A stable, proven stack.

We build technology that solves real needs. That is why we rely on proven components that work well together: AI models and infrastructure, operating system, web server, database and languages, chosen for each project.

01

AI infrastructure

Open-weight models such as Qwen, served with vLLM on our own NVIDIA GPUs, behind a LiteLLM gateway with keys, budgets and monitoring. Retrieval runs on PostgreSQL with pgvector and row-level security; agents reach tools through MCP and run long jobs as durable Temporal workflows.

  • vLLM
  • LiteLLM
  • Qwen · open-weight
  • NVIDIA GPU
  • PostgreSQL + pgvector
  • MCP
  • Temporal
  • Prometheus · Grafana

02

Open source

Linux server distributions are the foundation of a solid infrastructure: portable, able to run many instances at once and customisable without restrictions.

  • Linux · Debian
  • Apache
  • MariaDB
  • PHP
  • Docker

03

Microsoft technologies

When the context calls for it, the Microsoft stack offers built-in features, extended support and full integration with the ASP.NET applications common in business settings.

  • Windows Server
  • IIS
  • SQL Server
  • ASP.NET Core
  • C#

04

Languages and frameworks

The right tool for every need: web applications, interactive interfaces, heavy data processing, websites you can manage yourself.

  • C++
  • Python
  • TypeScript
  • Vue
  • Angular
  • React
  • WordPress

In production

The stack we use on our own products.

Open source · geospatial data

  • C++
  • C# / .NET
  • Vue 3
  • Docker
  • OGC
  • STAC
  • COG / COPC

Digital tour operator

  • WordPress
  • WooCommerce
  • Bootstrap 5
  • PHP
  • SEO

Let’s talk

Tell us what you want to build.

We will get back to you to arrange a first meeting, at our Milan office or online. If you prefer, call us Monday to Friday, 8:00–13:00 and 14:00–17:00 (CET).