Staff+ Software Engineer, Account Abuse (Machine Learning)
Core
Build machine learning systems to detect and stop account abuse and fraud at scale, ensuring fair allocation of computing capacity.
Role type
Staff+ Software Engineer (Machine Learning)
Builds
Real-time scoring systems, feature computation platforms, and automated model development tooling for abuse detection.
Domain
AI Safety / Fraud Detection / Account Integrity
Deliverable
production ML models
Required skills
Python, SQL, machine learning model training, production deployment, batch processing (Spark/Beam), workflow scheduling (Airflow), point-in-time correctness, tree-based models, unsupervised/clustering/graph-based detection
Preferred skills
Feature platform experience (Chronon/Feast/Tecton), stream processing (Flink/Kafka), AutoML, working with noisy/delayed labels, integrity/spam/fraud detection experience
Technologies
Python, SQL, Spark, Beam, Airflow, Flink, Kafka, Claude
Responsibilities
Build and operate feature computation platforms for training and real-time scoring; Train, evaluate, and deploy abuse detection models offline and online; Automate model development lifecycle using AI assistants; Implement backtesting, shadow deployment, and staged rollouts with drift monitoring; Partner with data scientists and policy teams to improve label quality.
Seniority
Staff+, hands-on IC with strategic impact
