Fraud & Risk Management Analyst (Payments, Neo-Banking & Crypto)

Fraud & Risk Management Analyst (Payments, Neo-Banking & Crypto)

Location:
Bengaluru (WFO) |
Function:
Risk & Fraud |
Type:
Full-time

Compensation:
Competitive base + performance-linked bonus + ESOPs based on performance|
Tools:
MacBook provided

Why FinFusion

FinFusion builds future-ready FinTech platforms—ZenPay (Neo-Banking & Global Payments) and Verafye (AI-powered Fraud & Risk). You'll join a hands-on, fast-moving team shipping real-time transaction monitoring at scale across cards, global bank rails (UPI/ACH/SWIFT/RTP/SEPA/), wallets, and crypto.

The Opportunity

We're hiring a
Fraud & Risk Management Analyst
with deep domain expertise in
banking or payments
(crypto exposure strongly preferred). You will transform raw data into actionable insights—
discovering patterns, designing fraud rules & business logic,
 and partnering with Data/ML to continuously
train and improve models
. Your work will directly lower fraud losses, increase approval rates, and protect customers across markets.

What You'll Do

• Transaction Monitoring & Detection
• Analyze production streams (web, mobile, API, POS) to uncover fraud patterns (account takeovers, mule networks, triangulation, friendly fraud, synthetic IDs, bot abuse, velocity/structuring, BIN/range abuse, chargeback rings).
• Design, simulate, and deploy
  risk rules
  (pre-auth & post-auth) with thresholds, risk scores, and step-up flows; own rule governance and drift checks.
• Crypto Risk
• Identify crypto-specific typologies (bridging/mixing, peel chains, sanctioned counterparties, wash trading signals, dusting); define controls for VASP/Travel Rule workflows.
• Model Partnership
• Translate patterns into
  features/labels
  ; curate training/evaluation datasets; review model outputs (precision/recall/AUC), and recommend thresholding & orchestration with rules.
• Decisioning & Orchestration
• Tune policy stacks (allow/deny/review), waterfall logic, step-up KYC/KYB, device/behavior checks, and 3-DS/SCA strategies to balance
  loss vs. conversion
  .
• Investigations & Casework
• Conduct in-depth analyses, create narrative memos, and recommend mitigations; support chargeback defense and dispute evidence packages.
• Compliance Alignment
• Partner with Compliance on
  KYC/KYB, AML/CFT, STR/SAR
  triggers, sanction screening, GDPR/India DPDP Act/PCI-DSS hygiene.
• Impact & Reporting
• Own and publish weekly dashboards:
  BPS fraud rate, authorization rate, false-positive rate, manual review rate, recovery %, chargeback rate by scheme/rail
  .

What You'll Need (Must-Haves)

• 5–10+ years
  in Fraud/Risk within a
  bank or payments
  company (issuer, acquirer, PSP, gateway, or neo-bank).
• Proven track record
  designing and shipping fraud rules
  and
  tuning risk engines
  in production.
• Hands-on with transaction data and
  advanced SQL
  ; comfortable with Python/R for analysis (no heavy ML coding required, but strong analytical rigor is essential).
• Working knowledge of
  card & bank rails
  (auth flows, MCCs, BIN ranges, 3-DS/SCA, chargebacks/disputes, UPI/NPCI constructs, reconciliation).
• Familiarity with
  crypto risk
  signals and Travel Rule/VASP concepts.
• Strong grasp of
  metrics & trade-offs
  (loss vs. conversion, precision/recall, latency).
• Excellent written analysis: incident memos, runbooks, and executive summaries.

Nice to Have

• Exposure to tools/concepts like device fingerprinting, bot mitigation, graph/link analysis, velocity stores, case management systems.
• Experience with real-time data stacks (Kafka/Kinesis), feature stores, or SIEM; basic knowledge of MLOps/experimentation.
• Prior work with sanctions screening, name matching, and geolocation/IP intelligence.

What You'll Work With

• Platforms:
  ZenPay (Digital & Gateway), Verafye (Fraud & Risk)
• Data:
  Real-time transaction streams across cards/UPI/bank rails/wallets + blockchain signals
• Team:
  Risk Engineers, Data Scientists, Product Managers, and Platform Engineering

Success Looks Like (First 6 Months)

• Ship
  2–3 high-impact rule packs
  cutting targeted fraud typologies by
  30–50%
  with
  ≤10%
  incremental false positives.
• Improve
  auth/approval rate
  by
  100–200 bps
  via smarter step-ups and thresholding.
• Establish a
  fraud intel operating cadence
  (dashboards, alerts, playbooks, post-mortems).