Technical and strategic writing for risk engineering teams at payment processors, neobanks, and BNPL providers.
Card networks and acquiring banks are increasingly asking payment processors to explain their fraud detection logic. Here is what model explainability means in practice, and what you should be prepared to document.
Read more
Card testing attacks cost payment processors millions in chargebacks and ban risk. Here is what the attack pattern looks like — and the specific velocity signals that detect it before damage is done.
Read more
A blocked legitimate transaction costs roughly $32 in lost revenue, customer support, and goodwill damage. Most fraud teams optimize for recall but ignore the full cost of precision. Here is the math.
Read more
Buy-now-pay-later platforms saw fraud loss rates increase in 2024 as bust-out and loan stacking patterns matured. This is what the current threat landscape looks like for BNPL risk teams.
Read more
Synthetic identity fraud — where fraudsters blend real and fabricated data to create new identities — is now the fastest-growing form of financial fraud in the US. Here is what the signals look like and how to catch them at origination.
Read more
Fraud scoring at p99 latency above 150ms starts affecting authorization rates and user experience. But accuracy degrades when you skip features to save time. We break down how to navigate this tradeoff in production.
Read more
More payments volume flows through Atlanta-based infrastructure than most people realize. NCR, Global Payments, and a new generation of fintech builders have made Atlanta a serious center of gravity for payment risk engineering.
Read more
Account takeover attempts leave detectable traces in session behavior, device patterns, and transaction velocity — if you know what to look for. Here are the six signals with the strongest predictive value in our production data.
Read moreTechnical posts and industry analysis for risk engineering teams. Infrequent, never promotional.