• 151

    Projects submitted

  • 109

    Published projects

  • 47

    Comments left

  • 66

    Number of votes

  • 52

    days until voting ends

← Back to list

KALIBR: platform for ALM risk assessment, analysis and management

  • Project manager on customer side

    Elena Bulanova

    VTB Bank (PJSC)

    Managing Director, Balance Sheet Risk Analysis Service

  • Category

  • Nomination

  • Goals

    1. To create a single parameterizable full-cycle system for calculating and analyzing ALM risks, providing flexible configuration of calculations, custom scenarios and adaptive presentation of results without complex and lengthy modifications of the calculation core, with calculations on both static and dynamic balance.

    2. To create an end-to-end analytical contour for factor analysis of risk metrics and data quality control: from business checks and detail configuration to explaining the causes of changes in indicators in the required dimensions.

    3. To ensure sustainable dual-contour operation: the production contour for reliable regulatory analytics, the laboratory contour for rapid testing of hypotheses, new risk parameters and ad-hoc tasks, with controlled transfer of changes to production.

  • Timeline

    march, 2025 — march, 2026
  • Project completion year

    2026

  • Project scale

    120000 human-hours
  • Results

    A single digital platform for ALM risk management has been created, uniting the key processes of the bank's balance sheet management — from data preparation and control to scenario modeling, factor analysis and visualization of results.

    On the import-substituted ALM Calculator, static and dynamic balance sheet modeling, flexible parameterization and a dual-contour architecture have been implemented, allowing new hypotheses to be tested without risk to regulatory reporting.

    The risk block, Treasury and business lines received a single analytical space for assessing management decisions, optimizing the risk-return ratio and improving financial result sustainability.

    The project laid the technological foundation for the transition from disparate calculations to proactive balance sheet management. Automation, data transparency and reproducibility ensure scalability of the solution and reduced operational risks.

  • Project uniqueness

    The system was implemented for VTB Group with its broad product line and numerous optional features and deal terms.

    The platform:

    • Enables flexible management of calculation assumptions and independent configuration of over 150 calculation parameters and analytical dimensions via the interface, without IT involvement for each new management request

    • Enables hypothesis testing and shows not only the final metric but also the reasons for its change: data scope, product dimensions, multi-scenario capability

    • Provides a parameterizable analytical contour for risk assessment and analysis of risk metric changes using dynamic balance sheet modeling

    • Enables visualization of risk metric calculation results and flexible dashboard configuration

    The solution reduces the "black box" effect in risk analytics. The project is fully implemented on an import-independent technology stack.

  • Software used

    The system is designed and manufactured in the Russian Federation on a modern technology stack:

    • Application layer: Backend — Java 21 + Spring Boot 3; Frontend — ReactJS

    • Infrastructure and data: container platform — Kubernetes; transactional DBMS — PostgreSQL; analytical DBMS — ClickHouse; long-term data storage — S3 + Apache Iceberg

    • Integration and processing: integration bus — Apache ActiveMQ Artemis; distributed computing — Apache Spark; process orchestration — Apache Airflow

    • Monitoring and analytics: BI and visualization — Apache Superset; monitoring and observability — Prometheus + Grafana

    The architecture ensures scalability, fault tolerance, centralized processing and storage of large data volumes, and independence from proprietary foreign platforms.

  • Solution from the Global CIO catalog

    The project does not use solutions from the Global CIO catalog

  • Project implementation complexity

    Such a complex project required a team of experts from different fields, with many challenges to address:

    • Complexity of data collection and calculations: information quality, fragmented data sources, complex behavioral models and metrics, dynamic changes in calculation and reporting methodology. This is not a "one-time calculation" system but a production IT function for critical risk types covering the entire balance sheet, with many products, clients, currencies, scenarios and dimensions.

    • Creating a convenient, flexible and understandable system with a user focus.

    • Ensuring reliable regular operation while enabling rapid changes, testing of new parameters and visualization of results.

    • Implementing functionality for VTB Group: complexity is driven not by the number of formulas but by the number of integrations, calculation options, data, performance requirements, controls, releases and the need to independently verify results.

  • Description

    KALIBR was created to solve a fundamental ALM analytics challenge: to reduce the time between a change in the economic scenario and understanding its impact on the bank's balance sheet. The bank needed not another data mart of indicators, but a single environment in which calculation, analysis and management decision form a continuous process.

    The project created a parameterizable calculation and analytical platform for balance sheet risk management. Its complexity lay in combining more than 20 data sources and calculation systems, flexible configuration of risk parameters and scenario assumptions, factor analysis and interactive analytics. Calculations use Russian software registered in the ERRP — the "ALM Calculator". The platform supports liquidity and interest rate risk calculations, dynamic balance sheet modeling, net interest income forecasting and behavioral models. More than 15 dashboards, data marts and analytical dimensions form a single space for the risk block, treasury and finance units.

    The key principle of KALIBR is parameterizability of risk analytics. The user can independently change over 150 parameters, create unlimited scenarios, select detail levels and analytical dimensions, test business hypotheses and see not only the final metric but also the factors behind its change. This reduces dependence on calculation core modifications and shortens the path from a new management question to an answer.

    The architecture follows a dual-contour model. The production contour ensures reliable regular operation, control and stability of approved analytics. The laboratory contour enables rapid testing of hypotheses, new risk parameters, scenarios and models before their controlled transfer to production. This allows the methodology to evolve without risk to regulatory processes.

    A separate layer ensures data quality and transparency: end-to-end business checks, unified preparation rules and reproducibility of calculations reduce manual processing errors. Apache Superset serves as a self-service analytical layer: users explore data in various dimensions and receive personalized reporting without contacting development for each new request.

    Thus, KALIBR moves balance sheet risk management from disparate procedures to a single managed digital contour. For the business, this means faster analysis of scenario impact on liquidity and interest rate risk, greater transparency of change drivers and lower operational risk; for IT — a single scalable architecture instead of a set of separate solutions. The platform helps the bank assess consequences faster, manage risk more accurately and balance stability and profitability.

  • Project geography

    The project was implemented for VTB Group, including the parent bank VTB, subsidiaries in the Russian Federation and the international network.

Comment

Only authorized users can comment.
We suggest you or register.

  • Customer

    VTB Bank (PJSC)

We use cookies for analytical purposes and to deliver you the best experience with our website. Continuing to the site, you agree to the Cookie Policy.