Developing data processing pipelines using machine learning models and integrating them into server and desktop applications. Also,training and researching machine learning models for the aforementioned pipelines.Relevant experience:• Developing a multi-model ensemble-based recommendation system for online commercial services, capable for automatic adjustment to the given service’s product catalogue (PyTorch, scikit-learn, Pandas,DVC, Flask);• Creating a speech enhancement (denoising), echo suppressing and speech detection application / software module in multiple programming languages (Pytorch, TensorFlow Lite, DTLN, C++, Java );• Developing a video conference assistant and building a video processing pipeline for it (PyTorch, TensorFlow, MediaPipe, ArcFace and RetinaFace models from InsightFace Solution, OpenCV);• Developing an outsorce solution for controlling the effectiveness of employees using people detection, face detection and face verification models (Triton, ONNX, YOLO V8, OpenCV, Docker, Docker Compose, RedisJSON);• Creating an automated offline deployment system for a video processing pipeline of a financial organization;• Implementing an attention control system based on telephone in hand detection in an outsource video conference assistant solution (YOLO V7, MediaPipe, TensorFlow).