Data Scientist

DATA SCIENTIST

ABOUT YOU:

This role is a member of the Data Science team, which is focused on developing advanced analytic models to understand consumer behaviors. The incumbent works directly with various business groups across the organization (Yearbook, Scholastic and College) to understand needs for advanced analytics and insights and recommend solutions to those needs.

• Support. Develop presentations around insights and tell compelling stories with the data, deliver results to constituents at different levels: strategic, tactical, operational and across business units (Yearbook, Scholastic, College).

• Update and Maintain. Integrate and prepare large, varied datasets, both internal consumer behaviors as well as external and 3rd party data sources such as consumer demographics, to gain a holistic understanding of consumer behaviors and preferences.

• Develop. Build and maintain predictive and machine learning models by using techniques such as logistic regression, gradient boosted trees, random forest and recurrent neural networks. Build and maintain unsupervised learning algorithms such as consumer segmentations using techniques such as k-means and KNN. Create and maintain recommender systems such as next best action, product affinities, likeliness of upsell at checkout, etc. Develop and maintain systems to continuously monitor model performance, such as model and feature drift.

• Advise. Oversee the implementation of the models into production AWS/Snowflake environments – ensure models are scoring properly.

• Strengthen. Analyze complex consumer data such as purchase transactions, demographic/psychographics, website behavior and email marketing data. Dig beneath the surface of a problem to ascertain the “Why” in the backend data, uncover the true reasons data patterns and skews may be showing up, and convert those findings into clearly understandable, data-backed recommendations, to influence future member engagement strategies.

• Help. Recommend campaign test designs and experiments to test hypotheses, such as A/B testing, champion/challenger testing, etc. Collaborate with the data engineering team to gather the right data from the right systems, as well as make recommendations for continuous data improvement and cleanliness pipelines.

• Typical/expected % of overnight travel. None to Minimal.

• Experience. At least 3 years’ relevant experience, including, but not limited to:
  Statistical modelling and advanced machine learning techniques, including supervised and unsupervised methods, experience in cloud-based ML/AI technologies – preferably in the Snowflake and AWS environments, programming experience in Python and/or R (both preferred) and SQL, experience using big data (billion+ rows) across numerous systems and platforms, specifically in spotting and fixing standard data anomalies seen at this scale, experience in presenting & communicating complex projects to non-technical teams.

• Education. Undergraduate degree in Computer Science, Machine Learning, Applied Statistics, Mathematics or a relevant Engineering discipline. Preferred masters level degree in a quantitative discipline such as computer science, mathematics, machine learning, applied statistics or relevant engineering discipline.

• Technical Skills. Understanding of big data principles and technologies with working knowledge of data integration using established methodologies and technologies.

• Attention to Detail. Ability to balance hands-on coding and modelling (existing and new models) on a frequent basis with being able to summarize findings and present insights to non-technical teams. Continuous Integration and Delivery (CI/CD) experience, preferably in Snowflake/AWS environments.

• Collaboration & Drive. Team player with a collaborative mindset, always prioritizing impact for the business. Innovative mindset – relentless curiosity of big data and its ability to predict consumer behavior. Preferred experience in a retail setting and familiar with typical advanced analytic algorithms used, such as consumer segmentation, consumer lifetime value, propensity to purchase models and recommendation engines.

• We care…