We develop state-of-the-art machine learning systems that can identify patterns and make unprecedented predictions in various domains. We utilise advanced neural network architectures normally reserved for academia and apply them to commercial applications.

Thanks to self-optimising algorithms (AutoML), we create highly accurate prediction systems that do not require expensive data specialists to operate. Our solutions offer sophisticated feature sourcing and feature engineering. Using contextual data and advanced time-series processing, we identify obscure patterns not evident in the original source data. Our solutions are tried and tested on the stock market.