Filtr uses AI to scan 1000+ jobs and finds postings that perfectly matches your resume

Uber’s newly formed AI Security team is building the foundation for dynamic, data-driven security systems and evolving Uber’s Zero Trust Architecture (ZTA) to be more risk-adaptive across authentication and authorization

Requirements

• 3+ years experience building and deploying ML models in production, with hands-on work in feature engineering, training, and evaluation.
• Proficiency in Python and ML frameworks (PyTorch, TensorFlow, or similar).
• Strong foundation in ML algorithms: tree-based models (XGBoost, LightGBM), classical methods (logistic regression, SVMs), and exposure to neural networks (CNNs, RNNs, Transformers).
• Ability to analyze business/security requirements and support translating them into ML use cases.
• Experience with risk, fraud, anomaly detection, or security-related ML systems.
• Familiarity with large-scale data/infra systems (Kafka, Hive, Spark, Flink, Pinot).
• Exposure to handling challenges such as imbalanced data, feedback loops, or iterative retraining.

Responsibilities

• Support framing business and security problems as ML tasks.
• Build and iterate ML models that enable risk-adaptive, real-time decisions.
• Engineer features from Uber’s risk systems, logs, and contextual signals.
• Deploy and maintain ML pipelines in production, ensuring reliability and scalability.
• Collaborate with senior engineers to integrate ML into Uber’s authentication and authorization systems.
• Build models and features, and take them into production
• Translate business and security needs into concrete ML problems

Other

• Bachelor's, Master's, or Ph.D. degree in a relevant field (not explicitly mentioned but implied)
• Strong communication skills and ability to work cross-functionally with infra, risk, and security teams.
• Ability to work in an office setting, with at least half of work time spent in the assigned office
• Must be eligible to work in the US
• Must be willing to participate in Uber's bonus program and other types of compensation