Pavlo Mospan

- Responsible for the conception, development and productive integration of an AI system for the automated processing of large amounts of audio and text data - Implementation of a cloud-native end-to-end ASR system for the automated extraction of relevant information from audio data - Fine-tuning and integration of LLM models, including prompt engineering - Implementation of MLOps workflows - Development of API-based backend components (REST) for integrating GenAI functionality into product architecture

- Development and elaboration of an Object/Person Detection and Tracking ShowCases - Web crawler construction - Data extraction, parsing - Feature Engineering - Working with large datasets

- Detecting multiple objects on the video - Testing and improving the model - Built asynchronous Pub / Sub pipeline for object detection project

Aufbau eines Auto-ML-Systems basierend auf Reinforcement Learning und Bayesian Optimierung zur Erkennung der erforderlichen Hyperparameter für das Training eines Roboters Unterstützung beim Robotertraining im Unity mit ML Agents Multiagenten Reinforcement Learning System im Projekt Smarte Batteriesteuerung für die Produktion

Analyse von Texten mit Naive Bayes Classifier Verwenden von NLP-Methoden für die Textverarbeitung Data Cleaning und Feature Engineering