Staff Data Scientist

General Motors•Warren, MI

9d•$160,000 - $246,000•Remote

About The Position

Turn complex business questions and high-value data into trustworthy, production-grade machine-learning solutions that improve decisions, automate work, and create measurable business impact across Sales, Service, Marketing, and Global Markets. This is a hands-on Staff Data Scientist role for an experienced individual contributor who can move seamlessly from business problem framing and analytical discovery to feature engineering, model development, production deployment, and continuous improvement. The role combines deep technical expertise with strong business judgment, helping teams adopt rigorous, interpretable, and reusable data-science practices at scale.

Requirements

Bachelor’s degree in Computer Science, Data Science, Statistics, Mathematics, Engineering, or a related field; advanced degree preferred.
8+ years of professional experience in data science, machine learning, applied statistics, or a closely related discipline.
Demonstrated experience taking machine-learning solutions from problem definition and proof of concept through production deployment and ongoing operation.
Strong proficiency in Python and SQL, including experience with production-quality code, testing, version control, and documentation.
Strong hands-on experience with common data-science and machine-learning libraries such as Pandas, NumPy, scikit-learn, PyTorch, TensorFlow, or equivalent technologies.
Experience with feature engineering, model evaluation, experiment design, statistical analysis, and communicating results to nontechnical audiences.
Experience deploying models through APIs, batch pipelines, notebooks-to-production workflows, or comparable production patterns.
Practical understanding of MLOps, including experiment tracking, model versioning, data and model monitoring, drift detection, retraining, and release management.
Experience working with large-scale data platforms such as Databricks, Spark/PySpark, cloud data warehouses, or equivalent technologies.
Demonstrated ability to operate independently, make sound technical tradeoffs, and deliver in a fast-changing, cross-functional environment.

Nice To Haves

Master’s or PhD in Statistics, Computer Science, Machine Learning, Operations Research, Mathematics, or a related quantitative field.
Experience in automotive, sales, service, marketing, customer analytics, dealer analytics, warranty, incentives, forecasting, or other operationally complex domains.
Experience with causal inference, time-series forecasting, optimization, recommendation systems, natural-language processing, or generative-AI-enabled analytical workflows.
Experience with MLflow or comparable tools for experiment tracking, model registry, and lifecycle management.
Experience with Azure, Databricks, REST APIs, containerized deployment, CI/CD, and cloud-native data or ML services.
Experience defining model governance, responsible-AI controls, interpretability practices, or risk-based evaluation standards.
Experience quantifying financial impact and partnering with Finance or business leaders to validate value realization.
Familiarity with enterprise AI platforms, including Glean, Azure AI Foundry, Databricks, or comparable platforms.

Responsibilities

Translate ambiguous business problems into clear analytical objectives, modeling strategies, and measurable success criteria.
Develop, validate, and improve predictive, prescriptive, forecasting, optimization, classification, and segmentation models.
Select appropriate statistical and machine-learning techniques based on the business decision, available data, operational constraints, and expected value.
Apply advanced methods such as time-series forecasting, causal inference, experimentation, natural-language processing, and optimization when they are fit for purpose.
Define data requirements and partner with data engineering and business teams to establish reliable, well-documented data sources.
Build scalable, reproducible feature pipelines and reusable analytical assets.
Perform exploratory analysis, data-quality assessment, feature selection, and leakage detection to ensure models are based on sound data.
Work across structured and unstructured data, including customer, vehicle, dealer, sales, service, warranty, incentive, and operational datasets.
Establish rigorous evaluation frameworks that reflect real-world business outcomes, not only offline technical metrics.
Assess model performance, calibration, bias, interpretability, robustness, and operational fit.
Explain model behavior, assumptions, limitations, and recommendations clearly to technical and nontechnical stakeholders.
Design and analyze experiments, pilots, and champion/challenger approaches to validate value before broad adoption.
Package and deploy models as reliable production services, batch processes, or decision-support capabilities in partnership with software, data, and platform engineers.
Establish reproducible practices for dependency management, versioning, data lineage, experiment tracking, and model release management.
Design model monitoring for accuracy, data quality, drift, latency, availability, and business performance.
Define practical drift thresholds, automated alerts, retraining criteria, and service-level expectations for models operating in production.
Investigate production issues, identify root causes, and improve models and pipelines through structured iteration.
Collaborate with product leaders, business owners, architects, engineers, IT, Finance, and other partners to deliver end-to-end solutions.
Connect technical work to measurable outcomes such as revenue growth, cost reduction, productivity, customer experience, risk reduction, or improved operational decisions.
Balance analytical sophistication with usability, speed to value, maintainability, and adoption.
Lead the data-science workstream from concept through production and continuous improvement, maintaining clear documentation and delivery accountability.
Serve as a technical authority and trusted advisor on machine learning, statistical modeling, experimentation, and production data science.
Raise the quality bar for model development through reusable patterns, code reviews, documentation, testing, and reproducibility.
Coach data scientists, analysts, engineers, and citizen builders on sound modeling practices and responsible use of AI.
Help teams evaluate and use platforms such as Databricks, Azure AI, Glean, and other enterprise tooling when they accelerate delivery without compromising quality.
Share lessons learned, reusable components, and practical guidance across the AI Center and partner organizations.