Principal Applied Scientist
About the role
Overview Ads Fraud Detection sits at the critical intersection of Security and Machine Learning, defending one of the largest advertising ecosystems in the world. We build and operate advanced ML systems that identify and neutralize fraudulent web traffic in real time at massive scale, protecting advertiser ROI, publisher integrity, and user trust. Our mission requires us to push the frontier of high-precision, high-recall detection while adapting continuously to an increasingly sophisticated threat landscape. As a Principal Applied Scientist on the Microsoft Ads Fraud Detection team, you will lead the most complex and ambiguous fraud challenges: designing new anomaly detection approaches to surface emerging attacks early, building robust cross-signal correlation frameworks to define and expand fraud perimeters, and developing next-generation ML models and low-latency pipelines that stop attacks before they scale. A key expectation for this role is to architect scalable solutions end to end, ensuring detection systems remain performant, resilient, and cost-efficient as data volume, attack complexity, and business demands grow. You will also provide technical leadership across the organization by mentoring other scientists, guiding project direction, and raising the bar on scientific rigor and execution quality. You will also ensure that every solution meets production reliability requirements. Your work will directly influence platform capabilities. Microsoft’s mission is to empower every person and every organization on the planet to achieve more. As employees we come together with a growth mindset, innovate to empower others, and collaborate to realize our shared goals. Each day we build on our values of respect, integrity, and accountability to create a culture of inclusion where everyone can thrive at work and beyond.
Responsibilities
- Independently lead and execute multiple high-impact fraud detection initiatives, turning ambiguous problems into measurable business outcomes.
- Develop practical, deployable machine learning (ML) solutions that operate reliably at production scale, balancing detection effectiveness