Data Scientist (in English)

1. Cluster analysis of the nanopores in ceramics, development of ML model to predict the quality of surface polishing. 2. Proposed a new type of highly efficient X-ray gratings with shifted layer boundaries. 3. Designed and develop software for modeling the growth of multilayer structures. Developed the physical model of multilayer X-ray grating growth.

I programmed in Python and worked in JupyterNotebook and JupyterLab environments. I use libraries pandas, numpy, matplotlib, scikit-learn, catboost, keras, tensorflow. completed projects 1. Optimization of energy consumption in steelmaking. The main goal of the project to predict steel temperature and as a result to avoid an excessive consumption of electricity spent on steel heating. The prediction model was built with the help of Catboost ML library. The high quality of the model was confirmed by the values ​​of the MAE and R2 metrics. Project technology stack: jupyter notebook, python, scikit-learn, catboost, matplotlib 2. Assessment of the age of a supermarket customers. Age assessmenet was based on clients photographs in order to analyze the purchases of customers of different age groups, and control the conscientiousness of cashiers when selling alcohol. Age recognition from a photo was implemented using the Keras library. The developed model has high productivity and high quality of age estimation. Project technology stack: jupyter notebook, python, scikit-learn, keras, tensorflow, matplotlib 3. Prediction of the gold recovery rate from the gold-bearing ore. The goal of the project was to predict gold recovery rate basing on ore composition and techological process parameters. The scikit-learn library was used to build the prediction model. Project technology stack: jupyter notebook, python, scikit-learn, catboost, matplotlib.