Fraud Strategy Specialist - Payment Risk

Felix Technologies

46d•Remote

About The Position

You'll join the Fraud Strategy team at Felix — the team responsible for how we detect, decide on, and prevent fraud across every product, payment rail, and country we operate in. We sit at the intersection of Data Science, Product, Operations, and Risk: we turn fraud signals into decisions, design the controls that stop bad actors, and make sure good customers move through with as little friction as possible. As Felix grows into new rails, new products and new markets across Latin America, our job is to let the business say yes to growth without letting fraud scale alongside it. You'll own fraud performance across cards, ACH, and digital wallets (Apple Pay / Google Pay). Rules are a last resort here, not a first instinct. We want someone who looks for the fix in the product — a 3DS step-up, a hold, a change to the money-movement flow — before ever reaching for a new rule. You'll go deep on the mechanics of how money actually moves and where it breaks — return codes, settlement windows, tokenization, chargeback reason codes — and pair that with hands-on data work to find the vector, size it, and prove your fix worked. This role is critical to protecting millions of Latinos in the United States while keeping the payment experience seamless for good customers.

Requirements

Bachelor's degree in a quantitative or related field (Math, Economics, CS, Statistics, Data Science) — or equivalent practical experience.
4–6 years in fraud / payment risk, with demonstrated depth in payment-instrument and rail mechanics (this is the core of the role, not a nice-to-have).
Strong working knowledge of 3DS / step-up authentication and how to deploy it to reduce fraud without over-penalizing good customers.
Deep knowledge of payment systems and their fraud surfaces: cards, ACH (return codes, NACHA), wires, settlement timing, tokenized wallets, and chargeback mechanics.
Proven ability to reduce fraud through product and action design, not only rule-writing. You think in terms of vectors and levers, not rule counts.
Advanced SQL and Python, with the ability to work with large datasets and run complex analysis independently.
Experience designing and reading experiments (A/B testing) to measure control effectiveness and conversion trade-offs.
Understanding of machine-learning concepts and their application to fraud detection (development skills not required).
Experience with BI tools (Looker, Tableau, or similar).
Strong communication skills for both technical and non-technical audiences, advanced English, and a strong analytical curiosity to go deep into the data to find fraudsters.
These are the applicable requisites, although equivalent competencies in any of the above will also be considered.

Responsibilities

Own fraud performance for our payment instruments and rails. For each material fraud vector, diagnose the root mechanism and design the highest-leverage way to bring it down — biasing toward durable product and action-based controls over rule sprawl.
Use the full toolkit beyond rules: step-up authentication (3DS / SCA), holds, paused payments, dynamic friction, and product/UX changes — applying the right amount of friction to the right customer at the right moment.
Partner with Product and Engineering to embed fraud controls into the payment and money-movement experience, rather than bolting rules on top after the fact.
Bring deep, instrument-and-rail-specific expertise: card authorization and 3DS/SCA, ACH return codes and NACHA mechanics, settlement timing and the fraud windows it creates, tokenized wallets (Apple/Google Pay), and chargeback reason codes.
Design and run experiments (A/B tests) to validate that a control actually reduces the target vector — and quantify its cost to conversion and good-customer experience.
Conduct deep-dive analyses on payment fraud schemes, organized fraud operations, and emerging threats across rails to uncover risky behaviors and turn them into action.
Partner with Data Science to inform payment fraud models — providing domain expertise on vectors and feature engineering (model development not required).
Author and maintain decisioning logic in our orchestration stack (Sardine / Oscilar) where rules are genuinely the right tool — and retire rules that overlap or underperform.