Middle data scientist in Pershiy Ukrayinskiy Mizhnarodniy Bank, AT / PUMB

Qualifications and experience. Necessary: ​​Higher financial and economic, mathematical or information technology education At least 2 years of work experience in the field of Data Science/Machine Learning in banking, finance or information technology At least 2 years of Python programming experience ( including libraries for machine learning numpy, pandas, scikit-learn, pytorch, seaborn) Experience with SQL at least 2 years (including writing queries, procedures and query optimization) Experience with version control systems (github, gitlab) Skills of preparing different types of data for modeling Skills of learning classic machine learning models (classification, regression, clustering) Skills of interpreting machine learning models Presentation skills, including explaining complex concepts in an understandable way for a non-technical audience Desirable: Skills of preparing reports for monitoring the results of the model using PowerBI, Tableu, SSRSSkills for the deployment of machine learning modelsSkills for training machine learning models based on neural networksA/B testing skills - BigData skillsYour role: High-quality and timely solution of business tasks in the direction of data analysis using machine learning methods to obtain the maximum business effectRegular communication with by other divisions of the bank to set tasks with business based on predicative analytics for the growth of the bank's profit Downloading and preparation of data for the implementation of Data Science models for solving relevant business tasks Implementation, training and interpretation of Data Science models for solving relevant business tasks Presentation of obtained results and insights to business customers Creation of necessary reports , monitoring the effectiveness of models Support of models, generation of ideas for their further improvement Research of new models, methods and tools for solving Data Science tasks, demonstrations and lectures based on research results for the team