Personal details

Ristea N. - Remote data scientist

Ristea N.

Based in: 🇷🇴 Romania
Timezone: Bucharest (UTC+3)

Summary

I am a data scientist at Microsoft with a PhD from University Politehnica of Bucharest, specialized in Machine Learning and Signal Processing. I enjoy machine learning projects and I have been working as a data scientist in multiple companies by now. Due to the nature of my past research/work I have experience in several popular languages and technologies for data science, i.e. Python, MATLAB, PyTorch, Tensorflow.
I teach Machine Learning and Signal Processing Theory at University Politehnica of Bucharest for 3 years, where I guided several students for personal projects.

I have over 6 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.

Work Experience

Machine Learning Scientist
Microsoft | Nov 2021 - Present
Python
C
Machine Learning
Mathematics
Deep Learning
PyTorch
AI (artificial intelligence)
I work to develop the latest models for deep echo cancellation for Teams.
Data Scientist
Veridium | May 2019 - Oct 2021
Python
Git
Machine Learning
Cassandra
Signal Processing
Docker
Kubernetes
Deep Learning
TensorFlow
PyTorch
I brought contributions to a handling biometric approach based on Machine Learning models. I used handcrafted feature combined with deep features from neural models in order to attain the best possible results.

Education

University Politehnica of Bucharest
Doctor's degree・Machine Learning in Signal Processing
Sep 2021 - Jun 2024
University POLITEHNICA of Bucharest
Master's degree・Image Processing and Machine Learning
Sep 2019 - Jun 2021

Personal Projects

Complex Neural Networks for Earthquake Source and Magnitude EstimationIconOpenNewWindows
2021
Python
Git
Signal Processing
Deep Learning
PyTorch
AI (artificial intelligence)
In this project, I proposed a novel approach for estimating epicentral distance, depth, and magnitude directly from individual raw 3-component seismograms of 1-minute length observed by single stations. The proposed convolutional neural network-based method is able to handle complex-valued representations of the seismic data in the time-frequency domain by using dedicated convolutional and activation functions. The validation experiments were conducted over a publicly available and large database, STanford EArthquake Dataset (STEAD). This is part of a research paper published at IEEE Geoscience and Remote Sensing Letters, a top-tier journal in the geoscience domain.
Deep Learning Data Set Generator for Automotive Radar InterferenceIconOpenNewWindows
2021
Python
Signal Processing
Deep Learning
AI (artificial intelligence)
A data set generator for radar interference mitigation. This is a solution to the lack of publicly available data sets. I proposed a solution based on MATLAB and Python, which generates a custom number of data samples, which mimic real radar data. This project could be used to train deep learning models as well as classical algorithms. This is part of two research papers that were published at VTC-Fall 2020 and CVPR Workshop 2021.