Design and implementation of ML model ensemble identifying clients with the intention to leave the bank in the near future.
Artificial Intelligence & Machine Learning Consulting
We are a pool of experts from academia and industry, ready to help you inject state-of-the-art technologies into your company's bloodstream.
Co-Founders & Members
Machine learning research engineer with 9 years of experience in deep learning, computer vision, and time-series analysis. Holds a MSc in System Engineering and Informatics and a PhD in Machine Learning from Brno University of Technology. Expertise in high-performance multi-GPU computing, predictive modeling, and signal processing. Proven track record in both academic and industry settings, including post-doctoral research at the University of Vienna and the Research Institute of Molecular Pathology. Published in top journals including Cell Structure, Journal of Mathematical Imaging and Vision, Frontiers in Public Health, and Neural Computing and Applications. Author of FakET and Redistributor.
Lukas Vrabel finished his master's in Artificial Intelligence at the Brno University of Technology. There he continued to perform research and education, publishing in the area of theoretical computer science - formal languages and automata theory.
Later he returned to machine learning and data science at Seznam.cz - a Czech-based search engine and one of the few remaining local competitors to Google in the world. There he worked at multiple positions ranging from Data Scientist to Head of Research Department. He focused on solving the problems mostly in domains of computer vision, NLP, semantic understanding of documents, and search engine development.
He entered the startup scene by establishing and growing the Brno office for US-EU startup studio Central Europe AI. During his stay there, he also worked on machine-learning based real estate price prediction startup.
Lukas is currently working mostly as a consultant, helping various companies with building data science departments, teams, products, and projects.
Selected Success Stories & Collaborations
Churn Detection
Data Pipelines
Helping migrate, automate and scale data processing and training pipelines to a cloud architecture to prepare for serving more than a hundred of publishers.
Indoor Tracking
Design of ML models significantly increasing accuracy of real-time location system (RTLS) for indoor tracking based on ultra-wideband (UWB) technology.
Face Biometrics
Leading and coordinating a project for face recognition and authentication using deep neural networks on mobile phones.
Keyword Spotting
Prototyping a speech command detection algorithm for a noisy environment of the driver’s compartment of a diesel or electric locomotive. The emphasis was on accuracy of predictions while running the model on an embedded device.
Biomedical Systems
Signal processing and data science for the 1st Dept. of Neurology of St. Anne´s University Hospital Brno. Multiple research projects focused on quantitative analysis of hypokinetic dysarthria and dysgraphia caused by Parkinson’s disease.
Gunshot Detection
Design and experimental evaluation of a ML-based system detecting gunshots or explosions in the streets. The system was designed with the emphasis on integration into intercoms.
Insurance Process Automation
Coordinating and developing ML systems for S9Y. This collaboration resulted in automated claim processing and correspondence anomaly detection for the Czech branch of Generali.
Our Tech Stack
