Principal Data Scientist

Principal Data Scientist

Principal Data Scientist

Location:

Ideal candidates will be able to report to our Pulse Point location at 5650 University Pkwy, Winston‑Salem, NC 27105. This role requires associates to be in‑office 1 - 2 days per week, fostering collaboration and connectivity, while providing flexibility to support productivity and work‑life balance. This approach combines structured office engagement with the autonomy of virtual work, promoting a dynamic and adaptable workplace.

Please note that per our policy on hybrid/virtual work, candidates not within a reasonable commuting distance from the posting location(s) will not be considered for employment, unless an accommodation is granted as required by law.

Monday to Friday, 8:00 am to 5:00 pm

• Serve as the methodology-expert on the team and provide guidelines to team for applying appropriate algorithms or solutions to approach different data challenges.
• Coach junior team members on various data science techniques..
• Keep abreast of the latest enhancements and innovations within the areas of data science and big data processing.
• Lead the design and execution of predictive modeling projects to address specific business problems determined by consultation with business partners.
• Provide guidance to the team on working with data-sets of varying degrees of size and complexity including both structured and unstructured data.
• At the expert level, piping and processing massive data-streams in distributed computing environments such as Hadoop to facilitate analysis.
• Leads batch and real‑time model scoring to drive actions.
• Leads the development of proprietary predictive modeling algorithms to build customized solutions that go beyond standard industry tools and lead to innovative solutions.
• Leads communication and publish results and address constraints/limitations with business partners.
• Determines the continuous improvement opportunities of current predictive modeling algorithms.
• Proactively collaborates with business partners to determine identified population segments and develop actionable plans to enable the identification of patterns related to quality, use, cost and other variables that have a broad impact.

• Bachelor's degree in Statistics, Computer Science, Mathematics, Machine Learning, Econometrics, Physics, Biostatistics or related Quantitative discipline.
• Seven (7) years in any occupation with predictive analytics experience.

• End-to end predictive modeling and large-scale data preparation (ETL, feature engineering, training/validation, scoring) using Python, Apache Spark (PySpark/Spark SQL), PostgreSQL, and advanced SQL
• OS-level development, system control, dependency management, reproducible environments, and ML pipeline runtime setup using advanced Linux and Python environment tools (e.g., Anaconda/conda, pip/venv, airflow)
• Distributed ETL and processing for massive structured and unstructured datasets using Hadoop and Hive
• Enterprise cloud data access, storage, and compute for analytics and model execution using AWS
• Supervised ML development, tuning, and evaluation using scikit-learn and XGBoost
• Regression/GLMs, hypothesis testing, confidence interval interpretation, ANOVA, residual diagnostics, and evaluation methods (e.g., propensity score matching) using advanced statistical modeling and inference
• Customized deep learning model development using tools beyond standard tools (e.g., PyTorch)
• Parallelization and performance optimization for compute-intensive workloads using multicore/OpenMP concepts
• Code versioning, branching, collaboration, and release management using Git
• Developing, testing, and validating…