Machine Learning Engineer - Wallet Intelligence and Machine Learning

Summary

Are you motivated to protect users and their accounts while delivering the best possible customer experience? Come join the Wallet Intelligence and Machine Learning team, where we help secure users' digital lives across Apple's devices without sacrificing privacy. Machine Learning Engineers here build analytical solutions and think deeply about where they fit into a larger system, staying ahead of fraud and applying the best privacy-preserving and fraud-prevention methods available to make Apple products, and especially Apple Pay and Apple Wallet, the safest platform people can use. The On Device Insights team at Apple develops machine learning models that run directly on users' devices to protect them from fraud, holding themselves to an exceptionally high bar for privacy. As part of the Wallet Intelligence and Machine Learning team, you will help secure users' digital lives across Apple's devices — including Apple Pay and Apple Wallet — without sacrificing privacy. This is a mission-driven team that thrives on hard problems, healthy skepticism, and open collaboration.

Minimum Qualifications

Experience with machine learning methods such as classification, clustering, and anomaly detection. Strong programming skills in one or more languages such as Python, Scala, or Java. Experience processing and analyzing data at scale using distributed data or compute frameworks. Ability to communicate the results of analysis clearly and succinctly to a range of audiences. Experience delivering results on ambiguous, loosely defined problems, working with others. Rigorous analytical thinking, including the ability to question assumptions, reason through a problem, and justify a recommendation with sound evidence.

Preferred Qualifications

Experience deploying machine learning in resource-constrained or real-time environments, such as on-device deployment, model compression, or optimizing for inference budgets. Experience with distributed data and compute frameworks such as Spark, Ray, or Daft. Familiarity with privacy-preserving machine learning techniques. Background in fraud detection, risk modeling, or security-focused machine learning. Familiarity with iOS development. We're open to a range of specializations and are excited by candidates who bring a differentiating strength to the team, whether that's a research background, deep systems thinking, or expertise we don't yet have. Tell us what you'd add.

Description

We are looking for a Machine Learning Engineer to help develop and launch on-device technologies that keep our users safe, working closely with engineering, security, program management, and business partners. Our work is applied and pragmatic by necessity. Models must run in real time and in the background on the device without slowing down something as simple as an in-app purchase, which means designing within real constraints like model size, inference budgets, and memory. Because we often need to anticipate fraud rather than react to each new pattern as it appears, we have to be proactive and think ahead. This role is a chance to take ownership of a problem area, build a system-wide understanding of where our models fit, and apply your expertise in machine learning in an innovative and fast-moving environment. If you're energized by ambiguity, motivated by a meaningful mission, and the kind of person who digs beneath the surface and questions your own assumptions before forming a recommendation, we'd love to hear from you.

Responsibilities

Take end-to-end responsibility for translating customer and security needs into machine learning solutions, from framing the problem through feature engineering, model development, training, evaluation, and reporting. Design and deliver models that operate within real-world constraints, balancing accuracy against latency, model size, and on-device compute budgets so that protection never comes at the cost of the user experience. Build and share a system-wide understanding of where our models fit into the user journey and the fraud-risk journey, and use that understanding to anticipate problems rather than react to them. Uphold and advance a high standard for user privacy in everything you build. Partner across software engineering, security, program management, and business teams to define problems, align on solutions, and communicate results clearly to both technical and non-technical audiences. Share your thinking openly, welcome scrutiny of your own ideas, and build trust with the people you work with.