Since 2017 we have been successfully running and continuously improving a fully automated algorithmic trading based on time-series data analysis and forecasting by recurrent neural networks. Currently used by fully regulated hedge fund p-hat.
Our solutions are implemented with 3 key elements in mind
Our work is based on the latest developments in the fields of machine learning. Inspired by feasibility study or applying our own research the focus is always on bringing quick value to the client.
Close collaboration with the client is crucial, and we use a combination of two methodologies SCRUM and CRISP-DM for successful development. This approach helps to fully understand the domain of interest and design final as well as partial goals. We work on the partial goals to quickly bring value to the client while iteratively heading towards the ultimate solution.
We develop solutions which are independent of any environment, framework or technology. Each project defines requirements based on which we can design the integration process and select technologies to meet these needs, with an emphasis on data privacy and security.
ECG beats classification
In collaboration with one of our partners, we are developing an AI solution which will be able to detect abnormalities in heart functioning successfully. Solving such a task would help doctors to make a better diagnosis of cardiovascular disease which is the number one cause of death today.
We have developed a proof of concept for breast cancer detection from mammography screening images by solving complex tasks of detection, segmentation and classification using artificial neural networks for computer vision.
Medical records analysis
Help medical facilities to optimize, automate and streamline processes by analysing medical records using natural language processing methods.
Prediction of renewable energy production
Green energy is an inevitable future with its advantages and drawbacks. Our goal is to generate the most accurate prediction of the production to increase the benefit of such resources.
How it works
We do not come and go. We prefer long-term partnerships with experienced companies in different business domains.