Staff ML Engineer — Ad Predictions & Retrieval

Apple; Cupertino (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Apple seeks a highly skilled Machine Learning Engineer to join the Predictions group. You will build core ML models powering ad predictions and monetization across App Store and News platforms, focusing on scalable, privacy-first solutions.

You will work at the intersection of applied ML, deep learning, and retrieval systems, developing models to predict user interaction, optimize marketplace outcomes, and scale across billions of queries while exploring LLMs, RL, and representation learning for

Qualifications

• 8+ years of experience applying machine learning and modeling at scale.
• Deep experience with neural architectures (Transformers, DNNs, RNNs) and training pipelines (TF, PyTorch).
• Practical understanding of reinforcement learning and bandit-based optimization.
• Experience with high-volume data pipelines, A/B testing infrastructure, and performance measurement.

Responsibilities

• Design and implement ML models to improve ad predictions and CTR/CVR.
• Develop and optimize retrieval algorithms for large-scale search/retrieval.
• Contribute to deep learning, ranking signals, and LLM-based ranking.
• Work with distributed datasets to discover signals and improve models.
• Collaborate with engineering, data infra, and product teams to scale models to production.
• Design and run large-scale experiments to validate model architectures.

Skills

ML engineering

Python

TensorFlow

PyTorch

Reinforcement learning

A/B testing

SQL/Scala/Java

Distributed data pipelines

Education

Bachelor's in CS/ML or related field

MS/PhD preferred

Tools

TensorFlow

PyTorch

Job description

Apple seeks a highly skilled Machine Learning Engineer to join the Predictions group. You will build core ML models powering ad predictions and monetization across App Store and News platforms, focusing on scalable, privacy-first solutions.

You will work at the intersection of applied ML, deep learning, and retrieval systems, developing models to predict user interaction, optimize marketplace outcomes, and scale across billions of queries while exploring LLMs, RL, and representation learning for