Anti-Fraud Engineer (NYC / MIA)

About Crossmint

Global financial rails are undergoing a once-in-a-generation transformation. Instant settlement. Programmable. Agent-first. Crossmint is the infrastructure helping companies build for that future. We are the leading all-in-one stablecoin and wallet infrastructure platform, enabling fintechs, enterprises, and agentic platforms to integrate stablecoin rails with speed, compliance, and scale. Crossmint provides everything enterprises need to ship smart financial rails, including smart wallets, cross-chain stablecoin orchestration, on/offramps, token checkout, and more, all through a single developer-friendly API. Trusted by more than 40,000 clients including global leaders like MoneyGram, Western Union, and Paga to nation states like the Marshall Islands, Crossmint powers stablecoin flows that move billions from cross-border remittances, global payroll, to the world's first digital UBI program. MiCA-authorized, PSD2-licensed, and SOC2 Type II certified, Crossmint serves 150+ countries globally across 50+ blockchains. Backed by Ribbit Capital, Franklin Templeton, NYCA, First Round, and Lightspeed Faction. We're building the infrastructure for the era of programmable finance. Join us!

About The Role

We are hiring our first dedicated Anti-Fraud Engineer. Crossmint moves billions across remittances, payroll, on/offramps, card rails, and agent-initiated payments. Every one of those flows carries a fraud surface, and today the controls live inside each product. This role builds the decisioning layer that all of them use, and owns the numbers it produces. The role has a loss line attached to it. You will know what fraud costs us, which controls moved that number, and which good customers we blocked on the way. If you want a job where the impact is arguable, this is not it. You will report to the Engineering Manager for Identity, Compliance & Risk and work day to day with our Risk Operations analysts, Compliance, Legal, Security, and the product teams whose flows you protect. Crossmint operates under MiCA, DORA, SOC 2, and PSD2, so your controls carry both a commercial and a licensing consequence. This is a hands-on role. You will write the code, read the data, and take the pager.

Responsibilities

• Own fraud decisioning — Build and own the services that score and decide onboarding attempts, stablecoin transfers, on/offramp orders, and card transactions. Turn analyst judgment into production rules and models, with clear thresholds, versioning, backtesting, and a fast path to roll back. Build the signals those decisions depend on: device, behavior, velocity, counterparty, and graph relationships across clients and wallets. Measure every decision. Own catch rate, false-positive rate, and fraud loss per product line, and publish them without being asked.
• Detect and respond — Build detection for account takeover, mule networks, first-party fraud, synthetic identities, refund and promo abuse, and agent-initiated payment abuse. Cut the time from a new fraud pattern to a live control. Same day when the pattern is clear. Lead the engineering side of fraud incidents: contain the loss, find the root cause, and close the hole in the product. Participate in the on-call rotation.
• Give Risk Operations leverage — Build the tools our analysts work in: case management, queues, reason codes, and rule tuning they can do themselves. Decide what we build and what we buy across device intelligence, transaction monitoring, screening, and case management, then integrate it properly. Produce the evidence Compliance needs for filings, audits, and supervisory review as an output of the system rather than a manual exercise.
• Stay close to the product — Review design documents for new products and name the fraud surface before launch, not after the first loss. Work with client-facing teams when a client's own fraud problem arrives on our rails. Keep current on fraud tooling, onchain tracing, and the AI-native workflows we use across engineering.

About You

Must-haves

• 5+ years building production systems, with 3+ years focused on fraud, risk, or trust and safety.
• A track record at a company where fraud carried a real cost: payments, card issuing or acquiring, banking, lending, marketplaces, or a crypto exchange. You can tell us which losses you cut and how you proved it.
• Hands-on ownership of a rules engine or model-serving path in production, plus the data pipeline behind it.
• Strong programming skills in at least one modern language. TypeScript is ideal.
• Fluency in the metrics and the trade-offs between them: catch rate, false-positive rate, precision at threshold, and loss rate.
• Comfort with SQL and with running your own analysis. You look at the data before you argue about it.
• The judgment to size a control against the loss it prevents and the friction it adds.
• Deep familiarity with current AI and agentic tools, and a view on how they change fraud work on both sides.
• Bachelor's degree in Computer Science or equivalent professional experience.

Strong plus

• Experience with stablecoins, wallet infrastructure, or blockchain payment rails, including onchain tracing and address risk.
• Card scheme fraud experience: chargebacks, dispute and monitoring programs, 3DS, and SCA exemptions.
• Experience with KYC/KYB onboarding abuse and synthetic identity detection.