We design and build technology for companies and professional firms, from the Republic of San Marino: artificial intelligence, sensors and robots. It is led by Gian Angelo Geminiani, who has worked in this trade since 1997.
Or write to us: info@ggtechnologies.sm
Around the chief executive works a network of collaborators and technology partners across several countries: software architects, electronics designers, system engineers, chosen for the project in hand. It is how a small firm keeps electronics, robotics and artificial intelligence together: one skill at a time, picked for the project.
What makes that possible is that we do not start from scratch each time. Data acquisition, running the models and machine control sit under every project, solved once, in DigiSense®: our framework, a registered trademark. The work goes on your problem, not on the scaffolding.
Via Marino Moretti 23, 47899 Serravalle, Republic of San Marino. C.O.E./VAT SM29141.
DigiSense® is a registered trademark of G&G Technologies S.r.l.
The chief executive has been on the Italian ministry register of innovation managers since 2020, and completed the European EITCA/AI programme in 2025.
By Gian Angelo Geminiani, in Italian: Oltre le dimissioni (2023), on people who build software without being programmers; Come trasformare telefoni dimenticati in strumenti straordinari (2026), on reusing old phones; and the public manual Claude: la guida completa, which is updated with the product.
Software designed and developed entirely in Europe.
We design and build artificial intelligence solutions for companies and professional firms, and we see their adoption through from start to finish: training the people, choosing the architecture, development, integration into processes and compliance with the rules.
AI agents that write code, brought into your team's development process, on the models and infrastructure you choose. Costs are estimated before the investment.
Models run on your own workstations or servers, or in a hybrid architecture that masks personal data before anything is sent outside. Every project is set up against the rules that apply to it: the GDPR, the AI Act, Italy's AI law (132/2025) and, for high-risk systems, Legislative Decree 160/2026. How it works.
Agents with a defined scope of action, access limited to what is strictly needed, and a person's approval for sensitive operations. We also review the AI tools already in use in your company. The reasoning, in writing.
Repetitive work moves to AI, integrated into the management system (ERP), the other software in use and email. Results are measured against indicators agreed before work starts.
Your company archive becomes something you can question: every answer cites the document it comes from. We extract data from contracts, invoices and technical documents.
We work alongside management on priorities, investments, the choice of suppliers and the rules for how your company uses AI. Then we build what has been decided.
In “Sensors”: nine hundred emergency workers followed for twelve months with the Italian National Research Council and four universities, and an algorithm that reads the electrocardiogram in real time on a phone.
The European EITCA/AI programme in artificial intelligence, and the NVIDIA course for people who teach deep neural networks. Every certificate has a public verification link.
Podz.AI is our own personal AI workstation. It installs in a minute and runs for a thirty-day trial, with every feature: it reads your documents on your own computer.
Where the boundary of an AI agent falls, what separates a number from a measurement, how long a device lasts when nobody updates it: the questions that come before the technology.
Head of information systems, tasked with putting the new ERP into production. Then the founding of SkyLab, which assembled workstations for architecture practices, with the graphics cards chosen to run CAD.
Project manager and software architect, then CIO, then head of development and digital innovation manager for the group. Production processes, logistics, sales network, configurators. Among the things designed, machine learning in production before it was called AI. The first dates from the early 2000s and watched over a mobile-racking warehouse — the kind that runs on rails and opens one aisle at a time: before the aisle closed again, a camera checked it was clear, recognising people and obstructions in shed light that shifts from morning to evening. Later, machine learning applied to quality control.
At Food Science, a traceability platform for fresh-cut fruit and vegetables — washed and packed, ready to eat: RFID labels read on the products and an algorithm that walks back up the supply chain, showing each party only what concerns them.
Co-founded and led: it brings algorithms into industrial processes. In 2019 Confindustria Romagna picked it as the first San Marino technology partner in its SmartTech team.
Almost five years as CTO of a company that designs, patents and manufactures wearable devices for vital-signs monitoring. It is among the affiliations on both “Sensors” studies, and its devices were assessed in two clinical studies Geminiani collaborated on.
The company was founded in 2022 and brings that expertise onto DigiSense® and onto Podz.AI.
The core is small and the network around it is wide: collaborators and partners chosen for the project, across several countries. That is why the person who answers you is the one who designs, with no layers in between: the person you talk to is the same one who builds the system.
No. The office is in Serravalle, clients are in Italy too, and the collaborators are spread across several countries. The software is designed and developed in Europe.
San Marino is an independent state, outside the European Union. The practical question for your data is where it gets processed and who answers for it: in an on-premise installation the processing happens on your own servers, in your own country, and the one who answers for it is you — you can verify that from your own network traffic. Where a service of ours processes data on your behalf, the applicable law and the place of processing are fixed in the contract, before anything starts.
The person who does the designing. The reply comes, within one working day, from the same person who signs the code, the quote and the scientific papers. The first conversation is about whether we are the right people for your problem: if we are not, saying so quickly is faster for everyone.
Because our work sits under confidentiality agreements and we keep them: a firm that shows you other people's names will one day show yours. It is the same choice we propose for your data. What can be checked from outside is on this page, and we will gladly talk about projects by describing the problems, not the names.