Hi, I’m Ivan.

For years I worked as a developer at Cerence AI, the company that supplies voice assistants to the world’s leading car manufacturers. My job was to make sure the system truly understood what the speaker was saying, including the subtlest nuances of language. In this kind of work even a small change can have cascading effects on the performance of the whole system, so every intervention, however minor, had to be designed and tested with great care, in continuous, fast-paced work to meet client requests.

In recent years, the rise of large language models (LLMs, the technology behind tools like ChatGPT) has made the traditional systems that work relied on obsolete. At the same time, it has opened the door to far more agile and customisable solutions, able to bring AI into the everyday processes of any business.

That is why I chose to specialise in hands-on AI system development with Turing College, one of the most advanced international organisations in this field. It was a sharp change of approach: before, I worked inside a huge product alongside hundreds of other developers, always handling individual pieces assigned in a highly structured way. With Turing I learned to work the opposite way: identifying the problems to solve myself, defining solutions with complete freedom and following a project from idea to final delivery. No longer a single piece of a larger system, but the whole process, first-hand. During the programme I built demos and prototypes for a range of simulated business scenarios, tackling concrete problems and understanding where AI can realistically be used in companies’ day-to-day work.

The natural next step is to bring these skills and this new approach into the real world, and that is where LipaLab comes from.

AI tools are now within reach of even the smallest businesses, but there is often no time to work out where to start and how to use them safely. I have seen how often technologies that are already standard elsewhere arrive here years late. In the meantime, local businesses miss out on benefits they could already have: less time lost on repetitive tasks, processes that are easier to manage, decisions made with more information at hand.

There is another side to it: many people already use AI, but only occasionally (for example, opening ChatGPT to have it write an email or a piece of text), without realising how much greater the benefits can be when it is properly integrated into their processes.

LipaLab exists to close that gap. I work from Friuli-Venezia Giulia and want to work closely with local businesses and firms. I bring the skills I built working on complex systems and following the latest developments in AI, and turn them into solutions designed around the real processes of the people who will use them every day.

The name comes from “lipa”, Slovenian for linden: a tree with light wood that is easy to work and carve, and traditionally the place where the village community gathered. That is the idea behind the project: light, tailor-made solutions, designed to help the people who use them grow.