Projekt‐ & Berufserfahrung
11/2021 – offenTätigkeitsbeschreibung
I worked on the Teams audio quality. I was responsible of developing and shipping models for millions of customers.Eingesetzte Qualifikationen
C++, Data Science, Datenanalyse, Docker, Maschinelles Lernen, Microsoft Azure, Python, PyTorch
9/2019 – offenTätigkeitsbeschreibung
Published several research papers at top-tier conferences and journals.
Taught signal processing theory applications to bachelor's degree students.
Conducted a team of students in research projects for publications.
Research projects on automotive radar interference mitigation via deep learning.
Computer Vision, Data Science, Maschinelles Lernen, Python, PyTorch, Radartechnik, Signalverarbeitung, TensorFlow
6/2019 – 10/2021Tätigkeitsbeschreibung
Brought contributions to a handling biometric approach based on machine learning.
Used handcrafted features (e.g. HOG) combined with features from neural models (CNNs) in order to develop a robust biometric model.
Developed production pipelines for model productization.
Data Science, Docker, Kubernetes, Maschinelles Lernen, Python, PyTorch, TensorFlow
6/2018 – 6/2019Tätigkeitsbeschreibung
Contributed to the development of a convolutional neural network (CNN) model for an in-cabin emotion monitoring system for the automotive industry. The solution was presented at CES.
Implemented several generative adversarial networks (GANs) architectures (Cycle-GAN, WGAN, GAN, and SimGAN) in order to perform style transfer for expanding limited data sets.
Developed an embedded data acquisition system that was used to extend data sets by other teams.
Computer Vision, Faltendes Neuronales Netzwerk (CNN), Maschinelles Lernen, Mathematik, Python, PyTorch, TensorFlow
10/2017 – 9/2018Tätigkeitsbeschreibung
Published several research papers at top-tier conferences and journals.Eingesetzte Qualifikationen
Computer Vision, Data Science, Maschinelles Lernen, Python, PyTorch, TensorFlow
6/2017 – 10/2017Tätigkeitsbeschreibung
Developed projects for media monitoring (RTSP and H264) with algorithms for 2D image recognition (ORB, SURF, and BRISK).
Developed a C++ application for automated data extraction from identity cards.
Developed an OCR system for training license plate recognition. I used classical graph algorithms combined with computer vision methods.
C++, Computer Vision, Git, OpenCV, Python, PyTorch
I teach Machine Learning and Signal Processing Theory at University Politehnica of Bucharest for 4 years, where I guided several students for personal projects.
I have over 5 years of experience in coding. I love mathematics since collage, when I won multiple Olympiads and contests. Moreover, I have a pure passion about machine learning and algorithm.