Lead Data Scientist
Electronic Arts
Description & Requirements
The Office of the CXO drives effectiveness across the EA Experiences organization with a focus on excellence in business operations, comprehensive fan intelligence, plus internal technology and business solutions. We are horizontal connectors empowering teams across the Experiences organization with the strategic prioritization, investments, resources, data, insights, and technology required to accelerate business outcomes in service of our goals.
And we want you to join us. We are hiring a Lead Data Scientist for our CXO Data & Analytics team, reporting to the Director of Decision & Data Science. This is an individual contributor role.
Responsibilities:
- Mentor junior team members on best methods and new technologies to stay current in the industry
- Identify, source, examine, clean, visualize and prep datasets for statistical analysis
- Model datasets and perform model fit tests
- Build effective relationships with Product Managers and other partners to identify and execute new Data Science opportunities across CXO and EA at large
- Communicate modeling results in business-friendly language to partners and department leadership by building PowerPoint presentations
- Deploy models for production into AWS SageMaker and work with our IT department to identify the necessary data feeds so models can produce inference
- Work with our UX designers to decide how model results will be shown in an application
- Monitor model accuracy over time and keep an eye out for model drift; retrain model when accuracy drops below agreed-upon threshold
- Periodically report back to partners on model performance
- Assist in designing A/B tests, including determining hypotheses, lift and sample size calculations and runtime recommendations
Qualifications:
- 8 or more years experience in Data Science with solid experience in NLP
- Master's in Statistics, Operations Research or related field (we will consider a Bachelor's degree with significant relevant experience)
- Proven track record in executing large projects across multiple team members
Technical skills:
- Expert in at least one Statistical Programming Language such as R or Python
- Experience with relational and distributed databases (SQL)
Job-specific technical skills
- Expert in Data ingestion and munging, like extracting elements of a json object, feature extraction, stratified sampling and dimensionality reduction techniques (i.e. Principal Component Analysis)
- Expert in Exploratory Data Analysis including visualizations, summary statistics, identifying and imputing missing data and checking for data quality problems
- Expert in statistical concepts such as descriptive statistics, percentiles and outliers, central limit theorem, distance measures, probability theory, bayes theorem, continuous distributions, hypothesis testing and regression modeling
- Experience with unsupervised and supervised Machine Learning models (i.e. Decision trees, Naive Bayes, Clustering, Ranking, Classification) as well as Machine Learning concepts like boosting, dealing with imbalanced classes, train, test vs. validation set and overfitting
- Bonus: Experience with NLP concepts such as Named Entity Recognition, constructing Document-Term-Matrices, Topic Modeling, Text Classification and LLMs
- Bonus: Experience with Model lifecycle management concepts such as model bias and drift, constructing data Labeling jobs for collection of 'golden dataset', and model deployment, monitoring and decommissioning in AWS SageMaker
Job-specific soft skills
- Customer Centric – You put the customer/player at the center when designing new offerings and processes. You understand how decisions and work impact the player.
- People Analyst – You understand that your work contributes to improving people's experiences, not predict when a machine is going to break down.
- Learner – You are a regular on Stack Overflow, know how to take apart a Vignette and are always looking for a new project that allows you to put that cutting edge technique to good use you read about last month
- Documentation Expert – You know just the right level of documentation to help yourself and the person who takes over after you understand what you did, when and why
- Teacher – You are patient and can field questions from team members around better coding practices or statistics related problem
- Experiment Designer – You can create up an A/B test in your sleep, but you don't let a less than ideal A/B test set up stop you from finding some answers