Staff Applied Machine Learning Engineer - Fraud & Abuse
Core
Design, build, and operate production ML decision systems to reduce payment fraud, account takeover, identity abuse, and other adversarial activity across Block's financial services.
Role type
Staff Applied Machine Learning Engineer (Fraud & Abuse)
Builds
Real-time and batch ML decisioning systems, low-latency model serving, decision APIs, and product controls for risk management.
Domain
Financial services, fraud detection, risk management, trust & safety
Deliverable
production ML models
Required skills
Production ML system design, low-latency inference, feature pipeline management, model monitoring, incident response, AI-assisted operations, cross-functional risk tradeoff analysis
Preferred skills
Graph-based fraud detection, behavioral sequence modeling, entity resolution, human-in-the-loop workflows, regulated financial services experience
Technologies
Python, Java, Kotlin, SQL, TensorFlow, PyTorch, XGBoost, LightGBM, Kafka, Kubernetes, cloud infrastructure, data warehouses/lakehouses
Responsibilities
Build and operate real-time/batch ML systems for fraud and abuse prevention; Integrate multi-source signals into decision APIs; Own production lifecycle including data contracts, drift detection, and rollback; Develop feedback loops and AI-assisted triage workflows; Partner with compliance and operations to balance fraud reduction with customer access; Create reusable evaluation capabilities for internal tools.
Seniority
Staff, hands-on IC with strategic ownership