AI Merchant Category Risk Scoring Developer

Ilvix Works

Remote · $100k – $130k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

Merchant category codes tell you what kind of business a merchant claims to be. They do not tell you whether the business is a front for card testing, a high-chargeback operation in disguise, or a legitimate business operating in a genuinely risky category. Risk scoring that goes beyond MCC — incorporating transaction pattern analysis, chargeback velocity, and behavioral fingerprinting — is what separates intelligent merchant underwriting from rubber-stamp approvals. Our payment platform underwrites 22,000 merchants across 140 countries. The current merchant risk model produces scores that correlate with chargeback rates at 0.61 Spearman rank correlation. We are targeting 0.78 by adding temporal transaction pattern features that the current model ignores. You will own the merchant risk model: redesigning the feature set around behavioral patterns, improving score calibration across merchant categories that have very different baseline risk profiles, and building the monitoring that alerts the risk team when a merchant score diverges significantly from their baseline.

Fully remote.

Responsibilities

Redesign merchant risk scoring with temporal transaction pattern features
Improve score calibration across merchant categories with different baseline risk
Build behavioral fingerprinting features for merchant transaction pattern analysis
Implement monitoring alerting when merchant risk scores diverge from baselines
Design evaluation framework using chargeback rate outcomes as ground truth
Collaborate with compliance team on card network reporting requirements

Requirements

3+ years in risk modeling, fraud analytics, or ML with payments domain experience
Experience with behavioral feature engineering on transaction event data
Strong Python and SQL skills, familiarity with Spark for large-scale processing
Understanding of calibration and ranking metrics for risk scoring
Knowledge of payment industry concepts: MCC, chargebacks, card network rules