Lead Data Scientist (Personalization & Algorithm Evaluation)

04 Jul 2024

Vacancy expired!

Data Scientists on the Disney Streaming Machine Learning and Innovation team develop, maintain, and evaluate recommendation and personalization algorithms for Disney Streaming's suite of streaming video apps, notably Disney+ and Hulu. They specialize in applying advanced machine learning methods and analysis techniques (statistical modeling, experimentation, causal inference) to meet strategic product personalization goals, and constantly seek ways to optimize operational processes.

Responsibilities :

This is an Individual Contributor leadership role in the area of content recommendation. As a member of this team you will collaborate across Engineering, Product, and Data teams to analyze user behavioral data to better understand how people are interacting with our platforms and content recommendations in order to drive algorithm and product improvements. The role has a specific focus on the area of pre-production metrics for algorithm evaluation - responsibilities include codifying pre-production metrics for product features, closing the performance gap between offline and online evaluation, and designing and owning automated reporting tools and processes.

- Analysis and Algorithm Optimization: Perform deep dive analysis on product interactions and user profiles as they relate to algorithm output in order to drive improvements in key personalization metrics - Pre-production metric design and ETL: Own pre-production algorithm metric design by defining metrics and evaluation criteria and by establishing upstream ETL requirements for metrics calculations - Dashboard and Report Design and Automation: Establish a house reporting style and work with Data Scientists and Data Engineers to leverage automation and alerting tools - Partnership and Messaging: Partner closely with stakeholders on Product, Engineering, and Data teams to identify and unlock opportunities in the area of algorithm evaluation and testing, and manage messaging around pre-production algorithm performance

Basic Qualifications :
  • 5+ years of analytical experience using Python or R
  • 5+ years using SQL
  • 3+ years of experience performing analysis of user behavior as it relates to a technology product
  • In-depth understanding of machine learning algorithm design and evaluation
  • In-depth understanding of statistical concepts such as hypothesis testing, experimental design, causal inference, and counterfactual evaluation
  • Experience with data exploration and data visualization tools such as Looker, Tableau, etc.
  • Experience with cloud services in a production environment (particularly AWS)
  • Experience loading and querying cloud-hosted databases such as Snowflake
  • Ability to explain how models are used and algorithms behave to both technical and non-technical audience
  • Ability to gauge the complexity of machine learning problems and a willingness to execute simple approaches for quick, effective solutions as appropriate
  • Strong written and verbal communication skills

Preferred Qualifications:
  • Experience building and deploying full stack ML pipelines: data extraction, data mining, model training, feature development, testing, and deployment
  • Familiar with metadata management, data lineage, and principles of data governance
  • Familiarity with automated deployment and CI / CD tools
  • Experience with model evaluation (IR evaluation, recommender system evaluation)
  • Experience using Spark or Scala
  • Familiarity with Python development ecosystem and technologies such as Databricks and Spark

Preferred Education :
  • Master's or PhD in statistics, math, computer science, social science, or related quantitative field

Additional Information :

Location - New York, NY preferred but also open to US Remote for the right candidate

#DisneyTech

  • ID: #43816460
  • State: New York New york city 10001 New york city USA
  • City: New york city
  • Salary: USD TBD TBD
  • Job type: Permanent
  • Showed: 2022-07-04
  • Deadline: 2022-09-01
  • Category: Et cetera