Senior AI/ML Engineer

Senior AI/ML Engineer Department: Engineering Team: Machine Learning Location: San Francisco, United States Type: Full-Time **About the Role (Sr AI/ML Engineer : Not Data Scientist)** Core Security Engineering’s mission is to make the Uber production environment secure by default and provide industry leading products and services to all Uber's production services and infrastructure. We are focused on building both security primitives and end users products that help Uber engineers to secure their service, build trust, and advance security to enable our global business. We are responsible for providing and managing systems, services, and libraries to provide access management, and enforcement at scale. The scope spans across multiple verticals like service-to-service authentication/authorization, employee to system auth, customer auth. You’ll work on critical distributed services at a massive scale crafted with the best security practices at the forefront. You’ll be accountable for designing and implementing the AI/ML based solutions to continuously scale and operate such foundational security services. What the Candidate Will Do ---- 1. Translate business and security needs into well-defined problem statements and solve them with the AI-first mindset. 2. Develop, iterate, and productionize ML models that simplify access management and control. 3. Integrate ML systems into Uber’s critical systems (Identity, Access, Authorization). 4. Collaborate across Security, Risk, and Infra teams to deliver scalable, production-ready solutions. 5. Provide leadership by mentoring junior engineers, evangelize ML best practices, and help shape ML strategy within AI Secury. \-\-\-\- Basic Qualifications ---- 1. 5+ years experience in formulating ML problems from ambiguous business requirements, especially in risk, fraud, or security contexts. 2. Proficiency across a broad range of ML algorithms: tree-based models (XGBoost, LightGBM), classical statistical models (logistic regression, SVMs), and deep learning architectures (CNNs, RNNs, Transformers), with the ability to select and apply the right ap