Machine intelligence for finance
Building intelligence
for finance.
Every institution answers the same questions about risk — AML, KYC, fraud, exposure — and answers them four separate times. ELVNFI is building the standard that makes them one decision, and the infrastructure to learn across institutions without moving anyone’s data.
Built for Nigeria. Designed for Africa.
The standard
One API.Every risk decision.
Every institution answers the same four questions about the same customer — is this laundering, is this person real, is this fraud, how risky is this exposure. Almost nobody answers them in one place.
They are bought from four vendors, returned in four schemas, reconciled by hand and argued about in four different audit trails. ELVNFI is building the contract that makes them one decision.
{
"subject": "cust_4f19",
"decision": "review",
"aml": { "screened": true, "typology": "layering" },
"kyc": { "verified": true, "confidence": 0.97 },
"fraud": { "score": 94, "signals": 5 },
"risk": { "tier": "high", "model": "rng-2.4.0" },
"explanation": "/v1/decisions/d_8f21/why"
}
# four answers, one schema, one audit trailIllustrative response. Shape, not a released contract.
AML & sanctions
Watchlist and PEP matching, structuring and typology detection, and the alert that comes out the other side — as fields, not as a vendor-specific report.
KYC & identity
Verification status, document and liveness outcomes, and identity confidence expressed on one scale instead of each provider’s own.
Fraud
Transaction, account-takeover and mule-network signals scored in line, with the reason each one fired attached to the decision.
Risk assessment
Customer, counterparty and portfolio risk, versioned so a score from six months ago can still be explained to a regulator.
Federated learning
PlannedLearn from every institution.Move no one’s data.
Financial crime is a pattern between institutions. The evidence sits in banks that cannot legally — or commercially — hand each other customer records. So every institution learns alone, from the fraction of the network it happens to see.
Risk-NG is being designed so that models train locally and only model updates are exchanged. Raw records stay inside the institution and inside the jurisdiction. That is an architectural property, not a certification — it is on the roadmap, and it is not built yet.
Shared model
Aggregated across participants
learns the pattern · never sees the rows
Bank A
accounts · transfers
Data stays here
Bank B
cards · merchants
Data stays here
PSP C
wallets · payouts
Data stays here
Raw records never cross an institutional or national boundary.
Fraud & transaction monitoring
Collaborative models across banks, issuers and payment networks, so a fraud pattern that appears at one institution is recognised at the next one before it lands.
AML & sanctions screening
Joint models over correspondent banking networks, where laundering is visible in the flow between institutions and invisible inside any one of them.
Credit scoring & risk modelling
Regional banks, bureaus and alternative lenders building richer — and fairer — risk profiles than any of their own books can support alone.
Financial crime & cyber risk
Consortia identifying coordinated attacks that are only legible across several institutions at once.
Our first product
Risk, with context.
The first implementation of the standard: open-source risk intelligence for fraud detection and AML, built for the rails Nigerian institutions actually run on.
Risk isn’t a property of a transaction. It’s a property of behaviour.
Transaction monitoring
AML & screening
Behavioural intelligence
Case management
Built for Nigeria
Built for Nigeria.Designed to scale across Africa.
Financial infrastructure in emerging markets has its own transaction patterns, payment rails and operational realities — instant transfers that settle in seconds, agent networks that move physical cash, wallets that behave nothing like accounts.
A risk model trained on card-present fraud in another market does not understand any of it. Risk-NG starts by understanding those realities. Then expands globally.
Money movement
Bank
core banking
Wallet
stored value
NIP
instant transfer
Merchant
acceptance
Agent
cash-in / cash-out
Bank
settlement
Risk-NG waitlist
Be early.
Risk-NG is being built in the open. Join the waitlist to get early access, product updates and invitations to the first private releases.
No spam. Just product updates.