Sr. Data Scientist, ML Model Evaluation - Apple Media Products

Seattle, Washington, United States
Software and Services

Summary

Posted:
Role Number:200268855
The Apple Media Products Engineering team is one of the most exciting examples of Apple’s long-held passion for combining art and technology. These are the people who power the App Store, Apple TV, Apple Music, Apple Podcasts, and Apple Books. And they do it on a massive scale, meeting Apple’s high expectations with high performance to deliver a huge variety of entertainment in over 35 languages to more than 150 countries. These engineers build secure, end-to-end solutions. They develop the custom software used to process all the creative work, the tools that providers use to deliver that media, all the server-side systems, and the APIs for many Apple services. Thanks to Apple’s unique integration of hardware, software, and services, engineers here partner to get behind a single unified vision. That vision always includes a deep commitment to strengthening Apple’s privacy policy, one of Apple’s core values. Although services are a bigger part of Apple’s business than ever before, these teams remain small, nimble, and cross-functional, offering greater exposure to the array of opportunities here. Are you passionate about analyzing data patterns for large scale search engines? Our team provides insights through data that drive decision making for our engineering and product teams. As part of AMP, you will be a product specialist and analyze of a diverse portfolio of Apple Media Product human evaluations tasks. The team represents the user’s perspective as they vet new features being introduced to AMP. Your day-to-day responsibilities will include analyzing and creating innovative ideas to help improve AMP search and recommendations features, and reporting detailed issues to leadership, engineering, and product groups.

Key Qualifications

  • Experience working with systems and products that are powered by machine-learned models and an understanding of natural language or speech-enabled products
  • 5+ years of experience with model evaluation (IR system evaluation, recommender system evaluation)
  • Research experience, ideally published, in regards to evaluation methodologies, evaluation metrics (NDCG, MRR, recall, precision, F-score, Precision@k)
  • A demonstrated passion for data analysis, data-informed product innovation, building and shaping data culture
  • Proven experience with at least one data-querying language (SQL and/or Spark, etc.) as well as with a scripting language for data processing and development (e.g., Python, R, or Scala)
  • You use good judgment balancing art and science when visually communicating information (e.g. Tableau, Superset, ggplot, D3).
  • Capable of driving projects of varying sizes and scopes - some will take months and some weeks — and you will need to know when to dive deep
  • Deep understanding of the software development for machine-learned products and services, and how analytics and human evaluation data can help improve those products
  • Ability to manage complex relationships across multiple functions and establish strong partnerships. Outstanding communication and presentation skills, written and verbal, to all levels of an organization
  • Good product sense, with the ability to translate between product goals, business goals and technical requirements for data analytics
  • Self-motivated and proactive, with demonstrated creative and critical thinking capabilities

Description

Lead the product vision and project execution for offline evaluations for Apple Music through Search, Recommendations, and Siri. Partner with Product teams to define ML model evaluation KPIs (NDCG, MRR, Precision@k, etc.) and create dashboards to monitor the performance of our AMP products. Define the instrumentation required for yielding sufficient data in regards to sampling, test set creation. Champion an industry-leading privacy-focused strategy for product measurement, model evaluation, and analytics for Apple Music. Own and evangelize standard methodologies for product analysis, measurement, and instrumentation. Lead model evaluation design and management exercises. Interface with vendors to establish rater pools for your evaluations. Evaluate and improve the quality of critical metrics for key products by analyzing customer pain points. Empower others to understand and leverage data for decision making. Ensure the measurement of new features are designed and instrumented in every release cycle. Productize dashboards that help product teams to improve Search & Recommendations every day. Work with leadership to build a strategy that allows us to leverage our data to create automatic feedback loops that improve the AMP products.

Education & Experience

BS in Computer Science, Statistics or similar. M.S. preferred

Additional Requirements