Description
Our team is responsible for managing model risk (losses caused by model inaccuracy). We perform independent assessment of model quality and configure automatic monitoring of the model risk level.
In our work, we study and test modeling approaches for a wide range of financial tasks - from building time series forecasts to determining the sentiment of news in trading terminals. We work with Classic ML, DL, and Agents based on GenAI (GigaChat).
The focus of our activities is to improve the efficiency of models impacting the Bank's P&L and to develop tools that help assess model performance under various future scenarios.
Responsibilities
- research approaches to modeling and validating various model classes
- understand model structure, test their correctness, challenge the developer's approach
- automate validation tests, improve model quality monitoring
- participate in piloting the Validator Agent
- develop pipelines for data quality checks
- assess the impact of models on processes
- improve methods for model risk assessment and predictive analytics.
Stack: Python, Machine Learning, Deep Learning, time series
Requirements
- work experience from 1 year
- relevant technical education
- deep understanding of ML, mathematical statistics, and probability theory
- good command of main Python libraries for machine learning and data analysis
- understanding of time series, DL, finance, and risk management will be an advantage.
Conditions
- comfortable modern office near Kutuzovskaya metro station
- work format: full-time office
- annual bonus
- corporate gym and relaxation zones
- more than 400 educational programs from SberUniversity for professional and career development
- regular meetups and a developed DS community
- extended voluntary health insurance, preferential insurance for family, and corporate pension program
- flexible mortgage discount equal to 1/3 of the Central Bank's key rate
- free SberPrime+ subscription, discounts on products from partner companies
- referral bonus for recommending friends to the Sber team.
