Leonid Butyrev

• Built ETL pipelines for data enrichment and implemented recommendation algorithms and infrastructure for news recommendation using Python, Kubernetes, Elasticsearch, Google Cloud, FastAPI, and Scikit-Learn. Orchestrated pipelines with Prefect and incorporated model management with MLflow. • Implemented caching with Redis and utilized ElasticSearch for data indexing and search. Built CI/CD pipeline for automated deployment and implemented streaming and batch pipelines with Apache Beam to process data.

• Designed, implemented and evaluated information extraction system for unstructured text using Tensorflow and Python • Improved training time for existing models by 10X with increased task performance. • Presented state of the art developments in machine learning, best practices and current projects to technical and non-technical audiences.

Implementierte Deep-Reinforcement-Learning-Algorithmen für Kollisionsvermeidung und Pfadplanung von mobilen Robotern und Dronen. Resultierte in einer Publikation.

Machine Learning, Künstliche Intelligenz, Mustererkennung