Developed CV ML model for over-the-network realtime object detection and translation. The model is live in the Babbly app.Simplified (10x faster model) and improved (5% less RMS error) the ML model for spatial crime prediction in US neighbourhoods. Results published in Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning ModelImplemented a data cleaning and preparation pipeline for precipitation radar images. Data was later used for ML powered precpitation nowcasting solely from cleaned radar images.Worked on prospecting business leads using networking, LinkedIn, and cold emailing. Established CRM processes for rest of the business dev team. Writing business proposals and specifiyng project requirements with clients.

Led a team of interns on a global-scale Object Detection competition achieving top 15% team ranking. Presented results at the ECCV 2018 conference in Munich.Boosted the 900 class classificator model accuracy by more than 8%.Actively promoting the Data Science team by giving motivational and educational presentations, along with writing technical blogs.Research, implementation, and maintenance of cutting-edge Machine Learning projects in the fields of Computer Vision and Natural Language Processing.

AI powered Drone designed for Detection, Tracking, and Assault of miniature objects from high altitudeDevelopment and supervision of the Object Detection and Drone Controlling moduleAdapting SOTA Deep Learning architectures for execution on low-energy consuming computers

Produced multi-platform mobile application without bugsUpgraded NAO humanoid robot with advanced language characteristics using Natural Language Processing techniquesDevelopment of back-end and front-end web solutions, along with mobile applications