One platform, five layers, zero black box.
A new technology, breaking with existing tools. End-to-end industrialisation, from raw data to a defensible decision, for ALM and banking risk. Calculations stay scientific, deterministic, auditable. The AI works as a specialist orchestrator, never as a calculation engine.
Six letters, one architecture.
From raw data to a defensible decision.
Architectural principle: calculations stay scientific, deterministic, auditable. The AI works as a specialist orchestrator, never as a calculation engine.
EVE (economic value of equity), NII (net interest income), LCR (liquidity coverage ratio), NSFR (net stable funding ratio), solvency ratio, capital metrics and ratios, gaps, cash flows. Deterministic, versioned, auditable: the same result on every run.
Captures intent, chains engines and routines, reports in natural language, locally, in the chosen deployment environment.
IngestionData
ComputationEngineering
ReportingThe answer
ConversationalIntuition
KnowledgeLearning
Expert analysis, in seconds.
You define the method. RIALTO replays it at every reporting date: variance analysis between two reporting dates, native or custom routines, deterministic calculations and traceable analysis.
- Suggested blocks: according to the metric analysed, RIALTO suggests the relevant analysis blocks.
- Defined in conversation: the desired analysis can also be described in the chat: RIALTO composes the routine.
- Documented justifications for the supervisor and audit.
- Capitalised expertise, shared across teams.
Query, simulate, explain in natural language: stress and reverse stress testing.
Across all scopes, interest rate, liquidity and solvency: stress simulation, reverse stress testing, explanation of results, decision support and RAF threshold monitoring, deepening the analysis question after question.
Shocks, stresses and scenarios, on demand
A shock, a stress scenario or an assumption change is described in one sentence in the conversation, or configured once in the menu, then replayed across all metrics. Simulation goes down to the finest level of detail provided by the bank.
From threshold back to scenario
Identify the scenarios that lead to a regulatory or internal threshold being breached.
Changes understood
Break down a change between two reporting dates (EVE, NII, gaps, LCR, NSFR, solvency ratios, RWA), identify its contributing factors and the causes of a RAF threshold breach.
Full traceability
Every figure stays linked to the assumptions that generated it, and to the source data.
Decision support
Preset, user-configurable analysis routines. Goal-based simulation: for a set target, RIALTO proposes changes to the selected balance sheet items.
RAF alerts
Immediate notification when regulatory or internal thresholds are breached. Real-time monitoring of indicator status (Safe / Alert).
ALM and risk dashboards you can explore.
Interest rate, liquidity and solvency: dynamic output and interactive bridges.
Dynamic output
Immediate access to clear dashboards to steer key indicators, with no technical effort. LCR dashboard, EVE by scenario, gaps and survival horizon, solvency ratios and RWA (risk-weighted assets).
Interactive bridge
Visual breakdown of the drivers of change between two or more reporting dates: rate, volume and time effects; LCR, NSFR, gaps and RWA as multi-bridge. Every effect can be clicked and drilled down.
Regulatory and management reports linked to the calculation.
Regulatory reporting automation: instant generation of regulatory and management reports, interest rate, liquidity and solvency, with continuity between the data, the calculation and the exported report.
A fully customisable report
Free selection of the metrics to include, by domain (interest rate, liquidity, solvency), with their breakdowns, sensitivities and analyses. Choice of global scope and/or by currency, subsidiary, business line.
All regulatory reports, by jurisdiction
The regulatory reports of every domain, interest rate, liquidity and solvency (IRRBB, LCR, NSFR, ALMM, solvency), produced to the requirements of each jurisdiction (ECB, ACPR, NBB, CSSF, Bank Al-Maghrib, BCEAO, COBAC…), directly from the engine results.
Your data, recognised and mapped.
RIALTO recognises the business meaning of your data and proposes a mapping to the target reference, adapted to your internal formats. Your teams validate or correct it before integration.
- The validated mapping is reused in production.
- Faster integration, consistent classification across reports.
- Auditable regulatory mapping for the LCR.
Trace every agent action.
From raw data to the delivered file, each step retains its parameters, timestamp and results. Figures defensible before the supervisor: ECB, ACPR, CSSF, NBB, BaFin.
- Every operation identified, parameterised and timestamped
- Input parameters listed, calculation stages detailed step by step
- Duration and status per operation, exactly reproducible result
Scheduled production, a complete history.
Simplified production
Scheduling and execution of your monthly reports, fully configurable. On-demand indicators, for instance automatic EVE / NII sensitivities under your management options (BAU, prepayments / renegotiations…). Real-time monitoring of changes and indicator status.
Reports and contexts kept on record
Full access to the report history, regulatory and internal management, sorted by user, date and time. Calculation contexts are kept on record too: reopen a past reporting date to analyse it, query it or run live simulations on it.
Your references, queryable.
A regulatory and internal knowledge base, queryable in natural language: teams query the standards, summarise the requirements and access the reference passages.
- Document grounding: answers rest on the validated corpus and point to the reference passages.
- Auditability and citations: the AI does not generate standards, it returns existing knowledge in a structured way.
- Safety: structural reduction of hallucination risk through a response frame restricted to the bank's documents and the built-in ALM expertise.
A controlled ETL chain.
An intelligent ETL chain reads the bank's sources or standardised files, proposes the mapping to the target data model, validated by the business and reused at every reporting date, and identifies the missing data for regulatory reporting, metrics and analyses. Ingestion, calculations and reporting start as soon as extractions are available.
Dedicated connectors
Extraction from the bank's systems through dedicated connectors or standardised files. Freshness on demand: extraction at every reporting date, no more divergent versions.
Reusable mapping
Extractions are automatically converted to the RIALTO pivot format, validated by the business.
The AI does not access the data
It orchestrates the engines and routines that use them, then returns their results.
Access management: role-based access.
User control
Creation, deletion, activation or deactivation of accounts. Access quotas tracked in real time. Direct administration of active members and invitations.
Centralised sign-in
Administrator: global system management, account monitoring, access to user-generated documents. User: direct access to the analysis tools and the AI chat.
See RIALTO reason over your numbers.
We simulate a full analysis with your data types and the scenario of your choice.