Hiring Fraud Engineers: The Complete Guide

# Fraud Engineer

**Location:** Remote (US) · **Employment Type:** Full-time · **Level:** Mid-Senior

## About [Company]

[Company] is a leading online marketplace connecting buyers and sellers across 50+ countries. We process $5B+ in transactions annually, and trust is the foundation of our platform. Every fraudulent transaction costs us money, damages user trust, and helps bad actors succeed.

Our fraud team is the front line of defense. We protect legitimate users from account takeover, prevent payment fraud, and stop bad actors before they can harm our community.

**Why join [Company]?**

- Work on fraud at meaningful scale—$5B+ GMV, millions of transactions
- Adversarial ML challenges—fraudsters are smart and constantly adapting
- Real business impact—every improvement saves real money
- Join a 15-person fraud team within a 400-person engineering org
- Series E company, profitable, industry leader
- Remote-first culture with competitive compensation

## The Role

We're looking for a Fraud Engineer to help us build the next generation of fraud detection and prevention systems. You'll work across the full fraud stack: real-time scoring, ML models, rule engines, and investigation tools.

This is adversarial work. Fraudsters are intelligent, well-resourced, and constantly evolving their tactics. You'll need to think like an attacker to defend like an expert. When you build a new detection, they'll find a way around it. When you close one avenue, they'll open another. The game never ends.

The ideal candidate combines ML skills with security thinking. You understand precision/recall tradeoffs in a world where false positives hurt users and false negatives lose money. You're comfortable with extreme class imbalance and evolving attack patterns.

## What You'll Work On

### First 30 Days
- Onboard to our fraud stack, data pipelines, and detection systems
- Understand our fraud types, attack patterns, and current defenses
- Ship your first fraud improvement (new feature, rule, or model tweak)

### First 90 Days
- Own a fraud detection area (payment fraud, account takeover, or abuse)
- Build new ML features or improve model performance
- Contribute to our real-time scoring system
- Work with operations team to understand edge cases

### First Year
- Lead fraud initiatives that measurably reduce fraud losses
- Architect improvements to our detection and response systems
- Develop new fraud signal sources (graph analysis, device fingerprinting)
- Mentor teammates on fraud engineering best practices

## Objectives of This Role

- Reduce fraud loss rate by 15%+ while maintaining user experience
- Improve model precision without increasing false positive rate
- Reduce fraud review latency (faster decisions, less user friction)
- Develop new fraud signals that detect emerging attack patterns
- Build tooling that helps operations team work more efficiently

## Responsibilities

**Model Development (50%)**
- Build and improve fraud detection ML models (classification, anomaly detection)
- Engineer features from transaction, user, and behavioral data
- Handle extreme class imbalance and adversarial drift
- Evaluate model performance across precision, recall, and business metrics
- Implement model monitoring for performance degradation

**Real-time Systems (25%)**
- Design and maintain real-time fraud scoring infrastructure
- Build and tune rule engines for fast response
- Optimize latency while maintaining detection quality
- Implement feature stores for real-time model serving
- Handle high-volume traffic with low-latency requirements

**Analysis & Operations (25%)**
- Analyze fraud patterns, attack trends, and emerging threats
- Build investigation and review tools for operations team
- Create dashboards and reporting for fraud metrics
- Participate in fraud case reviews to improve understanding
- Collaborate with operations on feedback loops and labeling

## Required Skills and Qualifications

**Technical Skills**
- 4+ years of ML engineering experience
- 2+ years in fraud, trust & safety, or security domains
- Strong experience with classification and anomaly detection
- Experience with extreme class imbalance problems
- Real-time system experience (sub-100ms scoring)
- Strong SQL and Python skills for data analysis
- Understanding of precision/recall tradeoffs and business impact

**Domain Knowledge**
- Understanding of fraud patterns and attack vectors
- Experience with feature engineering for fraud detection
- Familiarity with rule-based systems alongside ML

**Soft Skills**
- Adversarial thinking—ability to think like an attacker
- Analytical mindset for investigating patterns
- Collaboration with operations and policy teams
- Comfort with ambiguity—fraud detection is never perfect

## Preferred Skills and Qualifications

- Payment fraud experience (chargebacks, stolen cards)
- Graph-based fraud detection (link analysis, network patterns)
- Experience with device fingerprinting and behavioral biometrics
- Account takeover detection experience
- Marketplace or fintech fraud background
- Rule engine design and optimization

## Tech Stack

**ML:** Python, scikit-learn, XGBoost, LightGBM, PyTorch for deep models

**Real-time:** Apache Kafka, Redis, custom feature store

**Data:** Snowflake, dbt, Airflow for batch processing

**Rules:** Custom rule engine, Drools experience helpful

**Infrastructure:** AWS (EKS, Kinesis, Lambda), Kubernetes

**Monitoring:** Datadog, custom fraud dashboards, PagerDuty

## Compensation and Benefits

**Salary:** $145,000 - $195,000 (based on experience)

**Equity:** 0.04% - 0.12% (4-year vest, 1-year cliff)

**Benefits:**

*Health & Wellness*
- Medical, dental, and vision insurance (100% covered)
- $100/month wellness stipend
- Mental health support through Lyra

*Time Off*
- Unlimited PTO with 15-day minimum encouraged
- 11 paid company holidays
- 16 weeks paid parental leave

*Professional Development*
- $2,500 annual learning budget
- Conference attendance (KDD, Merchant Risk Council)
- Internal fraud and ML knowledge sharing

*Financial*
- 401(k) with 4% company match
- Equity refresh grants annually

*Workspace*
- $1,500 home office setup allowance
- Remote-first, no required office days
- MacBook Pro and equipment of choice

## Engineering Culture

**How we work:**
- Data-driven decisions—every change measured against fraud metrics
- Fast iteration with careful deployment (canary, gradual rollout)
- Close collaboration between engineering and operations
- Regular fraud case reviews to stay connected to real attacks
- Blameless post-mortems when fraud gets through

**What we value:**
- Adversarial mindset—think like an attacker
- Balancing precision and recall for business outcomes
- Continuous learning—fraud patterns evolve constantly
- Respect for the operations team who handle edge cases
- Work-life balance (on-call is rotational and well-supported)

## Interview Process

Our interview process typically takes 2 weeks.

- **Step 1: Recruiter Screen** (30 min)
  Background discussion and role overview.

- **Step 2: Technical Screen** (60 min)
  ML fundamentals and fraud concepts with a fraud engineer.

- **Step 3: Fraud Deep Dive** (90 min)
  Technical discussion about fraud detection approaches, tradeoffs, and your experience. How would you detect a specific fraud pattern?

- **Step 4: ML Coding** (90 min)
  Practical ML exercise—feature engineering, model building, or evaluation. Focus on fraud-relevant scenarios.

- **Step 5: System Design** (60 min)
  Design a fraud detection system: data flow, model serving, rules, feedback loops.

- **Step 6: Team Interviews** (2 x 30 min)
  Meet potential teammates. Discussion about collaboration and adversarial thinking.

---

*[Company] is an equal opportunity employer.*