Data Scientist, HopSkipDrive

• 5+ years of hands-on experience building and deploying production quality predictive and/or prescriptive models. Will accept substitution of up to 2 years for relevant educational experience
• Strong data acquisition skills using SQL in modern technical stacks (PostGres, AWS, Snowflake)
• 5+ years working with open source data science toolkits including Python (numpy, pandas, scikit, PySpark) or R (Tidyverse/Tidymodels). Strong preference for candidates who can capably translate between Python and R. Will accept substitution of up to 2 years for relevant educational experience
• 2+ years experience in deploying models to production environments through APIs, or batch processes
• 3+ years experience in at least TWO of the following DS methods
• Supervised and Unsupervised ML (regression, classification, etc.)
• Econometric modeling (e.g. time-to-event/survival analysis)
• Reinforcement learning (e.g multiarm bandits for price or intervention optimization)
• Operations research (e.g. combinatorial optimization/ traveling salesman, linear programming)
• Demonstrated experience with reproducible data science environments including creation of virtual environments, library management, and version control using Git/Github
• Demonstrated experience writing up results using literate programming workflows (Quarto preferred)

Desirable

• You will leverage organizational data to generate meaningful insights that improve the experiences of our users, while also being able to improve the effectiveness and efficiency of core business functions
• Combine DS strategies, programming expertise, and theoretical understanding of techniques to deliver advanced ML solutions for problem solving across the enterprise
• Scope, explore, specify, test, and deploy, and observe accurate and stable models into production environments
• Utilize modern, open-source data scientific workflows (SQL, R/Python, Github) to create highly accurate or explainable predictive/prescriptive models
• Own models in production and re-train when necessary
• Demonstrate a product mindset to deliver portable components that can be deployed into existing products OR converted into stand-alone data products
• Partner closely with BI, Engineering, Product Management, DevOps, and business stakeholders to create business-aligned, stable workflows
• Communicate results and model specifications to appropriate stakeholders
• Mentor, train, and assist aspiring data science professionals