We are looking for a Data Analyst to join the AI Business Assistant development team within an online bank for entrepreneurs (SberBusiness).
The goal of our Assistant is to take over routine banking operations, help select services that are suitable for a specific user, and suggest possible development vectors for their business based on previous performance.
User interaction with the Assistant takes place through a chat, which is powered by LLM, RAG, MCP, and, of course, classic application services. The system semantically distinguishes hundreds of related business scenarios, extracts critical information, and connects specialized GenAI tools to perform operations.
Responsibilities
- Data modeling: designing the structure of data marts to ensure high read speed and ease of writing analytical queries
- ETL/ELT process development: creating and configuring pipelines for extracting, transforming, and loading data into data marts
- Orchestration and automation: setting up data mart update schedules using schedulers and ensuring automatic restarts in case of failures
- Performance optimization: tuning SQL queries, configuring partitioning, indexes, and data distribution to speed up reports
- Data Quality assurance: setting up monitoring and data checks for completeness, uniqueness, relevance, and compliance with business logic before they reach users
- Documentation: maintaining technical documentation, describing data sources, calculation logic, and data schemas in the Data Catalog
- Working with ad-hoc queries: quickly retrieving data, testing hypotheses, and analyzing incidents.
Requirements
- SQL: ability to write complex queries (with JOIN, subqueries, window functions), create stored procedures, and optimize performance
- 1+ year of work experience
- Python for data analysis and visualization (pandas, numpy, scipy, matplotlib)
- good knowledge of mathematical statistics, probability theory, and experiment design (A/B tests, statistical significance)
- research experience: hypothesis testing, results evaluation, formulation of actionable conclusions
- understanding of multi-agent systems or complex LLM pipelines (orchestration, call chains, RAG).
As an advantage:
- understanding the principles of recommender or ranking systems and their evaluation methods
- familiarity with the Big Data stack (Hadoop, Spark)
- experience with git, bash/linux, development tools (Jira, Confluence)
- experience in banking or fintech projects
- experience with vector databases
- experience with LLM frameworks for agent orchestration (LangChain, AutoGen, CrewAI, LlamaIndex)
- systematic prompt engineering experience: managing response structure (JSON), reasoning chains, context length.
Conditions
- comfortable modern office near Kutuzovskaya metro station
- hybrid work format
- annual salary review, quarterly and annual bonuses (bonuses depend on position and bonus system)
- corporate gym and lounge areas
- over 400 educational programs from SberUniversity for professional and career development
- adaptation program and manager support at the start (for Junior level vacancies)
- extended VHI (Voluntary Medical Insurance), preferential insurance for family and corporate pension program
- flexible mortgage discount equal to 1/3 of the Central Bank's key rate
- Prime subscription with the option to share with three close contacts
- reward for recommending friends to the Sber team.
