Dark Pools FinCrimes
Graph analytics for fraud, money laundering and compliance
FinCrimes resolves customers, accounts, devices, policies and claims into one graph, exposes the rings and laundering networks behind them, and scores every event against 28 explainable typologies in real time. Compliance toolkits serve commercial banks, insurers, and the central banks, regulators and financial intelligence units that supervise them.
Capabilities
From the graph to the filed report
Five capabilities on one detection, investigation and automation kernel. Each reads and writes the same party graph, so a ring found by analytics is already the subject of the case.
CapabilityGraph analyticsMule rings, staged-accident rings and synthetic identities look ordinary one record at a time.CapabilityReal-time detectionThe FinCrimes detection engine scores a single transaction or claim in milliseconds and a historical backfile overnight, using the same 28 typologies on both paths.CapabilityAML complianceFinCrimes gives a bank’s financial crime team one system from first alert to filed report: transaction monitoring, sanctions and PEP screening, queues with SLA clocks, case management with an escalation ladder, and regulator-format filings.CapabilityCentral banks and regulatorsReserve banks, financial regulators and financial intelligence units see what no single institution can: the same people and accounts appearing across many reporting institutions.CapabilityInvestigationsInvestigators in FinCrimes work each alert or case with the facts, the transactions, the money flow and the network side by side.LibraryTypology libraryAll 28 schemes with red flags, data signals and response steps.
Who it serves
Built for the institutions that move money and the authorities that supervise them
Commercial banks and payment providersTransaction monitoring, screening, mule-ring detection and SAR/STR filing in one workflow, with the evidence an examiner asks for.Insurers and SIU teamsClaims, policies and providers resolved into one graph, so staged-accident rings and provider collusion show up as networks rather than single claims.Central banks, regulators and FIUsNetwork analysis across reporting institutions, report intake and triage, and in-country deployment configured to the authority’s mandate.Mobile money and fintechSIM-swap takeover, agent cash-out and corridor structuring scored in real time across wallets, agents and remittance corridors.
Detection engine
Explainable detection, end to end
Every alert shows its work: the exact records, counterparties, amounts and metrics that raised it. Alerts flow through SLA-driven queues into full case management and exit as filing-ready SAR and regulatory reports, with an AI copilot assisting at every step.
- Real-time scoring API to approve, review or block before the money moves
- Entity resolution collapses shared phones, devices and IDs into fraud rings
- Sanctions and watchlist screening at ingest, on demand and at decision time
- Per-entity baselines and adaptive thresholds score activity relative to history
- Shadow scoring and backtesting against your analysts’ past verdicts
- PII token vault: identifiers tokenized, retention enforced and every detokenization logged
Two licensed workspaces
Banking and insurance, hard-walled and independently licensed
One detection, investigation and automation kernel. Banking and insurance data never mix, with isolation enforced on every query and API key.
Banking
Insurance
Select a bar to list its typologies.
Insurance workspace
Motor · Life · Medical · Property · SIUMotor claims fraud
Life & medical fraud
Property claims fraud
Platform capabilities
From alert to filing-ready report
Alerts flow through SLA-driven queues into full case management and exit as filing-ready SAR and regulatory reports, with an AI copilot assisting at every step.
Real-time detection engine
28 transparent, parameterized typologies score streamed events in milliseconds and million-record backfiles overnight, on one engine with versioned rulesets.
Sanctions & watchlist screening
Parties, providers and counterparties screened at ingest, on demand and inline at interdiction decision time, every hit raised as a reviewable alert.
Entity resolution & ring analysis
Continuous identity resolution collapses "distinct" customers into a governed party master, exposing fraud rings on an interactive network graph.
Alert queues & case management
SLA-driven queues route work by severity into an investigator workbench with the full evidence trail. One click escalates alerts into audited cases.
AI copilot & assessments
Fraud score, risk level, ranked risk factors, anomalies and recommended actions, with explanations a reviewer or regulator can read.
SAR & regulatory filings
Generate SAR-style filings directly from case evidence, plus branded DOCX and PDF investigation reports, ready for the regulator.
Automation playbooks
Auto-escalate, assign and dispatch interdiction recommendations over HMAC-signed webhooks, with a per-workspace kill switch. Automation never moves money.
Governance & PII vault
Sensitive identifiers tokenized with retention policies and access logging. Approval workflows and immutable audit trails govern every action.
Operations & governance
Built for regulated operations
Financial crime operations
Real-time transaction scoring
Millisecond risk scores with approve, review or block recommendations before money moves.
SLA-driven alert queues
Severity-routed worklists feeding an investigator workbench with the full evidence trail.
SAR & regulatory filings
Filing-ready reports generated straight from case evidence in DOCX and PDF.
Governance & integration
Streaming & batch ingestion
Streaming ingest API with schema versioning and quarantine, plus CSV and API backfill.
Rules studio & backtesting
Tune typologies against your own history and analyst verdicts before going live.
Automation with a kill switch
Playbooks act at machine speed but never move money, and pause per workspace instantly.
Resources
Learn the schemes, then measure your controls
FAQ
FinCrimes questions
FinCrimes covers 28 fraud and AML typologies across two licensed workspaces. Banking covers structuring, mule rings, velocity spikes, dormant-account reactivation, round-tripping, synthetic identities, card testing, stolen-card bursts, BIN attacks, SIM-swap takeover, agent cash-out laundering, mobile-money Ponzi schemes, corridor structuring and funnel accounts. Insurance covers staged-accident rings, phantom vehicles, multi-claim VINs, inflated claims, ghost policies, early-death clusters, over-insurance, provider collusion, phantom billing, doctor shopping, impossible billing days, arson patterns, repeat losses and inflated contents claims.
Yes. A real-time scoring API returns a 0–100 risk score with contributing signals, matched typologies and an approve, review or block recommendation in milliseconds, before a transaction is booked or a claim is paid. The same engine scores streamed micro-batches and full historical backfiles, and every alert ships with an explainable evidence trail.
Get started
Take the next step
Resolve every party into one graph, expose the rings behind the transactions and see every alert show its work.