Real-Time Fraud ML Engineer – Production-Ready Modelling

Unity Technologies; Mountain View (CA); Presencial; USD 179.000 - 232.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Equity awards

Incentive plans and discretionary-bon-

Descripción de la vacante

Unity Technologies seeks a Machine Learning Engineer to own the ML side of Offerwall fraud prevention. You will bring real-time anomaly detection and device risk models into production, build labeled fraud datasets, and modernize training, serving, and monitoring across the fraud ML stack.

You will collaborate with data engineering, backend, and product partners and own end-to-end ML development from data to deployment, ensuring measurements align with business impact and reducing fraud losses.

Formación

• Production ML experience with deployment, monitoring and iteration after launch.
• Hands-on fraud detection or anomaly detection experience.
• Low-latency serving and online/offline feature pipelines.
• Ability to build labeled datasets and evaluation methods.
• Strong Python and SQL; cloud ML platforms (SageMaker, Kubeflow, or KServe).
• Ownership, collaboration with backend, data, and product teams.

Responsabilidades

• Ship fraud models to production influencing payouts and blocking.
• Create and maintain a golden labeled fraud dataset with evaluation metrics.
• Design streaming and batch features feeding a Redis online store.
• Plan and drive ML training/serving stack evolution (SageMaker→Kubeflow/KServe).
• Investigate fraud spikes and derive new detection signals and features.
• Monitor models, write runbooks, and participate in on-call duties.

Conocimientos

Python

SQL

Real-time ML

Fraud detection

Feature engineering

Redis

Kubeflow

KServe

SageMaker

Kafka

Airflow

On-call

Collaboration

Herramientas

Redis

Kafka

Airflow

SageMaker

Descripción del empleo

Unity Technologies seeks a Machine Learning Engineer to own the ML side of Offerwall fraud prevention. You will bring real-time anomaly detection and device risk models into production, build labeled fraud datasets, and modernize training, serving, and monitoring across the fraud ML stack.

You will collaborate with data engineering, backend, and product partners and own end-to-end ML development from data to deployment, ensuring measurements align with business impact and reducing fraud losses.