Artificial Intelligence Development Company

From the very beginning, we have been integrating machine learning (ML) solutions to improve our clients’ experience and ensure account security. Having an abundance of data coming from various sources, we designed and developed a continual learning framework that keeps all models up to date and sends alerts in case of suspicious activity, quality check failures, or drift.

Contributions: Automated fraud detection, personalized marketing, financial forecasting, internal credit scoring

Having started from a naive keyword-based approach for recommender and search engines, we kept experimenting, testing out new features and ML architectures. This resulted in a huge improvement in key product metrics. We followed a rigorous approach to get the most from our clients’ explicit and implicit feedback and trained a multimodal neural network that produces a set of semantic candidates that are scored afterwards by dedicated ranking models.

Contributions: Multimodal neural network for candidate generation, learning to rank model, autotagging model, automated ML lifecycle

With the focus on providing actionable insights for the ever-changing energy market, our expertise helped the platform gain a competitive edge by using a modern ML and data stack. We built energy supply/demand forecasting models and intelligent agents that can leverage these insights to perform automated decision-making and devise an optimal policy for consumers and producers. The system is powered by various sources that produce a huge volume of data that is ingested and processed in a streamlined fashion.

Contributions: ML-powered analytics, automated trends and seasonality detection, demand forecasting model, reinforcement learning agent to act upon market change