Description

We are looking for an experienced data processing and analysis specialist to help us improve our communication strategies and enhance the effectiveness of marketing campaigns through the development of specialized solutions in data processing and machine learning. Our team is engaged in designing data marts, creating analytical models, and developing approaches for campaign performance evaluation.

Main areas of work:

• Ensuring the collection, storage, and updating of data in corporate data warehouses (ClickHouse, Greenplum, Oracle).
• Optimizing processes for calculating complex customer characteristics and forming user profiles.
• Analyzing experimental data and developing algorithms to predict customer behavior.
• Preparing reports and visualizations to support informed management decisions.

Responsibilities

• Designing and building data marts, including tables and attributes for campaign analysis and audience segmentation.
• Creating analytical infrastructure for conducting tests, building communication funnels, and managing control groups.
• Building classic ML models and econometric estimates of the impact of various factors on the effectiveness of advertising campaigns.
• Collaborating with marketers to improve business metrics through deep analytics and the implementation of recommendation systems.

Requirements

Data Engineering:

• Advanced proficiency in SQL: working with window functions, Common Table Expressions (CTE), nested queries, index configuration, and partitioning.
• Experience in designing relational database structures for campaigns and segmented audiences.
• Application of orchestration tools (Airflow): creating workflow schedules (DAGs), handling exceptions, using sensors and timeouts. 
• Using Python libraries for efficient processing of large data volumes.
• Integrating Python code with data sources, developing loading scripts.

Data Science (ML & Econometrics):

• Deep immersion in classical Machine Learning: knowledge of classification and regression methods (boosting, decision trees, linear models, clustering).
• Ability to develop customized features (feature engineering) for specific business tasks (predicting purchases, churn, optimal customer offer).
• Competent interpretation of results and detailed evaluation of model errors.

Expertise in Econometrics:

• Evaluating the causal impact of interventions (Causal Impact, Diff-in-diff, synthetic control groups).
• Ability to build uplift models and understand biases and hidden factors (instrumental variables).

Metrics, Control Groups, Experiments:

• Conducting experiments (A/B tests): Organizing sample randomization and checking the correctness of group allocation at SQL/Python levels.
• Conducting cohort analysis, calculating LTV, retention, conversion rate, ARPU, and other product metrics.
• Pre-experiment preparation and post-analysis (variance reduction methods like CUPED, bootstrapping, Bayesian approach).  

Soft Skills:

• Attention to detail in data analysis and anomaly detection.
• Excellent statistical thinking and a critical approach to decision-making.
• Creative approach to hypothesis formation and developing new data to improve model predictive power.

Conditions

• Hybrid work format, in a modern office in Moscow on Prospect Mira (Olympiyskiy 14, Diamond Hall BC)
• Favorable mortgage and preferential lending conditions
• Free SberPrime+ subscription, discounts on products from partner companies: Okko, SberMarket, Sber Eapteka, and others
• Voluntary health insurance from day one and preferential insurance for close relatives
• Corporate pension program
• Children's recreation and gifts paid for by the Company
• Company-sponsored training: online courses in Sber's Virtual School and unlimited access to the library, training in the Corporate University, workshops, meetups, and the opportunity to obtain new qualifications