Machine Learning Engineer - Icloud Anti-abuse

Apple Apple · Big Tech · San Diego, CA +1 · Software and Services

Machine Learning Engineer for iCloud Anti-Abuse team at Apple, focusing on building and deploying ML models for abuse detection (spam, phishing) at massive scale. The role involves the full ML lifecycle from data pipelines and feature engineering to model training, low-latency inference, and monitoring within distributed systems.

What you'd actually do

  1. Own the end-to-end ML lifecycle for abuse detection across Mail, Calendar, and Contacts: data pipelines, feature engineering, model training, deployment, and monitoring
  2. Build and maintain ML infrastructure that operates reliably at iCloud scale with low-latency, high-availability requirements
  3. Develop techniques to identify and score abusive actors and patterns at scale
  4. Analyze model performance, identify failure modes, and drive continuous improvement
  5. Partner with backend engineers and cross-functional teams in trust and safety, operations, and product

Skills

Required

  • 3+ years of hands-on machine learning engineering experience
  • training and deploying models in production
  • Strong programming skills in one or more production languages (e.g., Java, Scala, Kotlin, Go, Python)
  • Experience building and operating ML pipelines: data processing, feature engineering, training, serving, and monitoring
  • Solid foundation in distributed systems
  • Familiarity with classification, ranking, or anomaly detection techniques
  • Ability to drive projects independently from problem definition to production

Nice to have

  • 5+ years of ML engineering experience
  • Experience with abuse detection, fraud prevention, content filtering, or trust and safety systems
  • Expertise in NLP or text classification applied to email, messaging, or similar domains
  • Experience with streaming/real-time ML inference
  • Familiarity with techniques for scoring, ranking, or classifying actors and behaviors at scale
  • Understanding of privacy-preserving ML techniques and responsible data handling
  • Experience with email protocols (SMTP, IMAP) or messaging infrastructure
  • MS/PhD in Computer Science, Machine Learning, or a related technical field

What the JD emphasized

  • production ML models
  • low-latency serving infrastructure
  • massive scale
  • production-quality code
  • distributed system tradeoffs
  • abuse detection

Other signals

  • production ML models
  • distributed systems
  • low-latency inference
  • massive scale