Marketing

Spotting Duplicate Accounts and Fraudulent Signups

• Queue Design for Fraud Review: What a Human Needs Beside a Score
• Routing Signup Risk Alerts to the Right Reviewer
• Using a Model to Score Signup Risk from Device and Email Signals
• Using an AI Agent to Read Referral Notes and Flag Signups
• When to Use an AI Agent to Review Referral Signups Before Payout
• A Golden Set for Queue Routing Decisions
• A Regression Set for Fraud Scoring Prompts When Models Change
• A Regression Set of Device Signals for Agent-Scored Signups
• Building a Golden Set of Referral Signups for Regression Testing
• Calibrating a Fraud Score Against Reviewer Decisions
• Choosing the Threshold: Calibrating Signup Risk Scores
• Detecting Duplicate Accounts with Email and Device
• Normalizing Device and Email Signals Before the Model Sees Them
• Prompt Injection in Signup Fields: What a Scorer Ignores
• Routing Referral Reviews: Rules First, Model Second
• Setting a Cost Ceiling for a Fraud Scoring Agent
• Storing Regression Set Inputs and Expected Outputs in Postgres
• Using a Model to Score Referral Notes for Fraud Signals
• What a Referral Review Agent Costs per Thousand Signups
• Writing an Evidence Tag Taxonomy That Survives New Data