Verifex Shield

Stop fraud before money moves.

Real-time fraud decisioning for PSPs, wallets, fintechs and banks - with explainable verdicts, velocity rules, device intelligence, sanctions evidence and audit-grade decision records.

Start a Shield pilot View API example

Shadow-mode pilot

Evidence-backed decisions

Private deployment options

Shield decision

Low-latency decisioning target

delay

{
  "verdict": "delay",
  "risk_score": 87,
  "recommended_action": "hold_and_notify",
  "rules_triggered": [
    "ATO_SEQUENCE",
    "NEW_RECIPIENT",
    "DEVICE_MISMATCH"
  ],
  "capsule_id": "frd_cap_0391"
}

Event

Rules

Score

Verdict

Evidence

The problem

Fraud policies do not stop fraud. Live decisions do.

Most payment companies already know what suspicious behavior looks like. The problem is execution: rules live in documents, analysts see alerts too late, and auditors still need proof of why a transaction was allowed, delayed or blocked.

Rules stuck in documents

Risk policy lives in spreadsheets and PDFs while payment flows keep moving.

Manual review after damage

Analysts see alerts after funds moved, chargebacks arrived, or payouts cleared.

No evidence trail

Teams need to prove why a transaction was allowed, delayed, held, or blocked.

Fraud and compliance split

Behavioral fraud, sanctions evidence, and merchant risk often sit in separate tools.

What Shield is

A decisioning API for risky financial actions.

Your backend sends a transaction, session or payout event. Shield evaluates behavior, device, velocity, counterparty and compliance signals, then returns a verdict and recommended action.

Client App

Backend

Shield API

Verdict + Evidence

Backend acts

Where Shield sits

A clean fraud decisioning layer between events and action.

Event Ingestion

Normalize logins, card adds, payments, deposits, withdrawals, payouts, and disputes.

Enrichment

Attach device, counterparty, merchant, payment, and optional compliance context.

Rules Engine

Evaluate explainable policy rules with active, shadow, and disabled modes.

Velocity Store

Track bursts, repeated failures, shared devices, and fast cash-out windows.

Risk Scoring

Produce a deterministic score and level from rules, velocity, profiles, and signals.

Action Engine

Return a backend-ready action instead of leaving engineers to interpret a score.

Evidence Capsule

Store the decision, checked signals, fired rules, timestamps, and hashable record.

Case Management

Route review-worthy decisions to analysts with the evidence needed to act.

Reporting & Webhooks

Send outcomes, route notifications, and preserve feedback for tuning.

Admin Dashboard

Review decisions, timelines, rules, cases, and pilot performance in one place.

Decisions

Return actions, not just scores.

A risk score alone still leaves your engineers to decide what to do. Shield returns a recommended action your backend can execute.

Clear / Allow

Proceed without additional friction.

Monitor

Let the action pass while tagging the customer or merchant for follow-up.

Step-up

Ask for email, phone OTP, 3DS, or KYC refresh before continuing.

Review

Send the event to a human queue with evidence attached.

Delay

Slow down the action while more checks or notifications run.

Hold funds

Keep money from leaving until the risk team clears the case.

Block

Stop a high-confidence risky action before value moves.

Pause payout

monitor, step-up, or block

Evidence

tenant-scoped velocity counters retained

Evidence Capsule

Every decision needs proof.

Shield stores an audit-grade record of what was checked, which rules fired, which signal snapshot was used, what score was produced, and what action was recommended.

Designed for audit review.

event_id

verdict

risk_score

rules_triggered

signal_snapshot

list_versions

timestamp

hash

Hashable · Immutable · Auditable

Fraud + compliance

Fraud risk and sanctions evidence in one platform.

Because Shield runs on Verifex infrastructure, clients can combine behavior-risk checks with Verifex Core screening modules such as sanctions, PEP, KYB or counterparty screening when needed.

API

Built for backend integration.

request

POST /v1/shield/decide

{
  "event_type": "transfer.created",
  "customer_id": "cus_123",
  "amount": 100000,
  "currency": "AZN",
  "recipient_id": "rcp_456",
  "device_id": "dev_789",
  "ip_hash": "ip_abc",
  "timestamp": "2026-06-20T10:30:00Z"
}

response

{
  "status": "completed",
  "outcome": "delay",
  "risk_score": 87,
  "recommended_action": "hold_and_notify",
  "rules_triggered": [
    {
      "id": "ATO_SEQUENCE",
      "reason": "New device + new recipient + high-value transfer"
    }
  ],
  "capsule_id": "frd_cap_0391"
}

Shadow mode

Launch safely in shadow mode.

Shield can run silently before enforcement. It calculates what would have happened, without blocking real customers. Risk teams compare recommendations with real chargebacks, disputes and manual review outcomes before enabling hard actions.

Observe

Run Shield beside production flows and collect recommended decisions.

Tune

Compare recommendations with outcomes and calibrate rules before enforcement.

Enforce

Enable low-friction actions first, then hard actions after review.

Hosted, private cloud or local deployment.

Start with hosted Verifex infrastructure, or deploy a dedicated regional/private instance for data-residency requirements.

Deployment model depends on client legal and compliance requirements.

From first event to pilot in weeks.

Week 1

event schema + fraud scenarios

Week 2

technical MVP + shadow responses

Weeks 3-4

rules + velocity + dashboard

Weeks 5-8

pilot tuning + reporting + rollout plan

Build a fraud layer your risk team can explain.

Verifex Shield turns policy into live decisions, live decisions into evidence, and evidence into a pilot your risk and engineering teams can trust.

Start a Shield pilot Talk to Verifex