The creation of intelectual property is our company’s primary objective on the mid and long-term. We invest a great amount of our resources into R&D. The purpose is better defining and perfecting our product lines while keeping our technology stack in sync with core market needs and trends.



Software as a service emerged as a revolutionary commerical concept in the late 2000s as a child of the global IT giants. Global manufacturing industry followed shortly, with no more than 10 years later, hardware manufacturers starting to lease components and provide service subscriptions instead of just raw bare-metal sales. Industrial equipment condition monitoring and production QA is one of the major topics on this, still developing industry, which we are more than enthusiastic about taking part in.


VIBRAT.IO is our cloud and on premises solution for equipment condition monitoring. It uses a distributed microservices architecture having at its core a collection of deep learning algorithms trained for sensorial signal processing. At install-time our models start-off with a 95% prediction accuracy. We have specialized algorithms for processing vibration data and hyperspecialized ones for bearing monitoring. Our differentiating KPI is having achieved the point of zero configuration, whereas existing solutions require more than several product descriptive parameters and extra running environment data such as the mechanism’s rotational speed.


The VIBRAT.IO platform works:




We’ve invested 5 years into developing the means to create AI based products in a more scalable and easier manner. In the last decade deep learning algorithms became more than successful, nevertheless, individual localized efforts seemed rather uncorrelated and lacking the coreography which would allow for cummulative and aggregative results to emerge at a product level.

After a rather short research time, our goal became to set the grounds for factoring-in diversity under a single hood. We’ve thus created GLAS.AI® as a framework which is able to agreggate discrete DL algorithms through a proprietary communication protocol designed to allow for an orchestrated and holistic experience.


GLAS.AI® is modeled after the human brain. A fully data driven distributed runtime, it brings-in customization and configuration NRCs, only. A privacy by design architecture at core, it can be used to create fully GDPR compliant components.

We’ve mainly targeted mobility and automotive use cases, nevertheless, the applications’ range is indefinite:

GLAS.AI® is ideal for developing:

Our research on GLAS.AI® discrete deep learning algorithms is focused on:


Development SDKS for:

Contact us for the full company Overview