Download PDF


I am a self-motivated data scientist. I have experience in development and upgrading of library of predictive maintenance tools. Able to play a key role in analysis of problems and come up with creative solutions as well as producing methodologies for anomaly detection, high dimensionality and other problems, machine learning faced with. A quick learner who can absorb new ideas and can communicate clearly and effectively.

Work History


Data Scientist

Siemens LLC, Engineering Center for Data Processing and Analysis

R&D in the mobility division at Siemens with focus on rail industry data.

Technical tasks:

- Cleaning and exploring data;
- Development of machine learning algorithms, using a wide range of methods (Convolutional Neural Nets, Recurrent Neural Nets, SVMs, OneClass SVMs, Random Forests, Boosted Decision Trees, etc);
- Statistical Data Visualization and Analysis

Business tasks:
- Present findings to both internal and external clients, including board members. 


Intern Researcher in Machine Learning

Institute for Information Transmission Problems (RAS),

Development of a library of predictive maintenance by Laboratory of Data Analysis and Predictive Modeling / DATADVANCE . Solving specific problems that emerge in the analysis of data obtained from cooperating companies

  1. Skoltech/3D Geometry: Research work in the field of 3d models (the quality of the reconstruction etc). 
  2. Skoltech/RNNreg: Research on recurrent neural networks with regularization on the weights matrix 
  3. Skoltech/ASOD: Aerospace data classification with CNN.
  4. EZY/Big Data: Implementation of a specific type of distributed computations system for large dataset (Amazon EC2 etc)
  5. Visillect/RNN TS Classification: Research in the field automatic building of recurrent neural networks techniques for automatic development of classifier for specific type of time-series. Implementation of the algorithm for car pass detection at the check-in points on the toll road.  (ICMV'2016, Nice, France: "Automatic Construction of a Recurrent Neural Network based Classifier for Vehicle Passage Detection")
  6. DATADVANCE/Leakage detection: Implementation of the algorithm for trend detection in air-cooling system for airplane (One-class classification for time series)
  7. Airbus/PMToolsImplementation and addition of new data analysis techniques (in particular for time series module: implementation of automatic ARIMA parameter fitting, t-digest, time-series anomaly detection)



Quantitative Analysis Program

Moscow State University, Center for Mathematical Finance

One year program in quantitative methods in finance.


MSc in IT / Applied Mathematics and Physics

Skolkovo Institute of Science and Technology / Moscow Institute of Physics and Technology

Data Analysis / Predictive Modeling and Optimization 


BSc in Applied Mathematics and Physics

Moscow Institute of Physics and Technology, Department of Control and Applied Mathematics, GPA 4.78/5.00