Description
We are forming a team to create a new tool – «Feature Store».
The tool should significantly simplify the life of DE/DS and accelerate T2M for the development and industrial deployment of new models.
Responsibilities
- development of the tool's architecture for working with data within the development and implementation of machine learning models in the Risk Block (Feature Store)
- analysis of the Risk block's storage and populating the Feature Store database with metadata (description of data marts, tables, attributes, table join rules, etc.)
- testing the SQL query generator for collecting training samples (a key component of the Feature Store)
- performance analysis of generated queries
- testing the Feature Store interface from the perspective of ease of implementing user scenarios.
Requirements
- at least 2 years of experience as a business or system analyst for DWH
- good knowledge of database theory (Normal forms, cardinality of relationships, etc.)
- excellent knowledge of SQL, experience in writing complex analytical queries
- experience with Hadoop and pyspark and query optimization (preferable)
- experience in building physical models of corporate data warehouses/data models.
Conditions
- office near Tulskaya metro station
- work format: hybrid
- annual salary review and annual bonus
- corporate gym and relaxation areas
- more than 400 educational programs from SberUniversity for professional and career development
- extended voluntary health insurance (VHI), 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
