Artificial Intelligence


There are many confusions around the term "artificial intelligence". The media and advertisers heavily abuse the idea, and recent researches about how many institutions declaring the usage of AI do not use it at all. In a recent study (source: click) by MMC out of 2,830 startups in Europe that were classified as being AI companies, only 1,580 fit that description accurately!

In general, AI enables the software to perform difficult tasks more effectively by learning through training instead of following sets of rules. There are many fields of research, like deep learning which focus on particular types of AI or particular fields of applications. AI is important because for the first time traditional human capabilities can be undertaken in software inexpensively and in a bigger scale. AI can be applied to every sector in order to enable new possibilities and efficiencies. All of those achievements are possible thanks to the availability of huge training data sets and powerful hardware.


Where can you find AI?

Pentacomp Systemy Informatyczne S.A. is an independent Polish software house. Since 1995 the company has supported enterprises and offices in achieving of their targets and optimizing their operation with the application of IT solutions and services. It has been over 20 years since we have been building Business Intelligence solutions, Data Warehouse and Artificial Intelligence, regardless of the AI definition adopted.


  • predict expected lapse probabilities for policies (improving customer retention)
  • optimization of insurance settlement paths (bigger savings with quick path of insurance settlement, that fits the clients‘ needs; including analysis of pictures)
  • forecast for the new policies purchase (increase revenues)


  • forecast for a market (marketing campaigns) and the supply (audience measurement) and better usage of commercial time.

Life Sciences (Heathcare)

  • the fraud detection module that is built based on graphs models and scoring modules for a real big data warehouses (over 1 billion records). Correlation analysis between entities, patients and benefits, which are repeatedly. Moreover, the module detects the benefits that should not be handed over to specific patients and identifies the entities and people who basically are responsible for the abuses.

Industry and retail trade

  • Cross-selling and up-selling analysis which purpose is to increase the market and booster the effectiveness of sales of goods.

What do we aim to create soon?

Our expertise has been proved in the projects mentioned above. We build Artificial Intelligence solutions that are based on both Open Source technologies, for example Python (Scikit-learn, TensorFlow, PyTorch, Keras), R, H2O, as well as commercial tech stack, e.g. Azure Machine Learning, SQL Server ML Services, SAS Enterprise Miner, DataWalk.

Do you want to know what special solutions we aim to create in the future?

Please feel free to contact us: