Machine Learning Engineer - Ads Relevance & Quality

Apple Apple · Big Tech · Cupertino, CA · Software and Services

Machine Learning Engineer to join the Ads Relevance and Quality team, responsible for building intelligent systems to evaluate ad relevance, detect low-quality or offensive content, and optimize user satisfaction using applied ML, NLP, and content quality techniques.

What you'd actually do

  1. Design and implement machine learning models to evaluate and improve ad relevance, trust, and quality for user queries
  2. Build NLP and multi-modal models that detect offensive, unsafe, or policy-violating content at scale
  3. Develop methods for semantic query understanding, ads understanding, relevance scoring, and keyword-to-ad matching
  4. Collaborate closely with product and policy teams to translate content integrity standards into measurable ML objectives
  5. Work with large-scale, privacy-preserving datasets to discover and operationalize new quality signals

Skills

Required

  • Machine learning
  • NLP
  • Transformer architectures
  • TensorFlow or PyTorch
  • LLM fine-tuning
  • Evaluation frameworks
  • A/B testing
  • Python
  • SQL
  • Problem-solving
  • Communication

Nice to have

  • Scala
  • Java

What the JD emphasized

  • 4+ years of experience applying machine learning at scale in domains such as ad tech, content moderation, search ranking, or recommendation systems
  • Strong expertise in natural language processing, including offensive content detection, semantic matching
  • Experience with Transformer-based architectures (e.g., BERT, DistilBERT) and training pipelines in TensorFlow or PyTorch
  • Familiarity with fine-tuning Large Language Models (LLMs) for downstream tasks such as classification, content moderation, or semantic relevance
  • Familiarity with quality and fairness evaluation frameworks (precision, recall, coverage, policy alignment, etc.)
  • Hands-on experience with A/B testing, experimentation frameworks, and performance debugging in production

Other signals

  • ML models for ad relevance and quality
  • NLP and multi-modal models for content detection
  • semantic query understanding
  • large-scale, privacy-preserving datasets
  • offline/online experiments