Data Scientist - AI/ML Solutions PPCO

General Motors•Warren, MI

2d•Hybrid

About The Position

The Role General Motors is seeking a Data Scientist to join the Product Program Cost Optimization AI/ML Solutions team. This role develops and scales production-grade data products, predictive models, and AI capabilities that improve cost intelligence and support product and program decisions. The Data Scientist will translate ambiguous cost, engineering, purchasing, and finance problems into scalable technical solutions. They will own work across the full product lifecycle: problem definition, data preparation, modeling, application development, deployment, monitoring, support, and continuous improvement. This role requires demonstrated depth in production machine learning, data engineering, and modern AI applications. The successful candidate will build governed capabilities that convert complex engineering, supplier, manufacturing, and financial data into measurable improvements in cost decisions, speed, quality, and adoption.

Requirements

Bachelor’s degree in computer science, data science, engineering, statistics, mathematics, operations research, physics, or a related quantitative discipline.
Five or more years of relevant experience, or three or more years with a related master’s degree, in data science, machine learning, ML engineering, data engineering, applied analytics, or a related role.
Advanced Python proficiency, including experience developing modular, testable, maintainable, and production-quality code.
Strong SQL skills, including designing, querying, integrating, and optimizing relational data.
Demonstrated experience building scalable data pipelines, transforming large datasets, and establishing data-quality controls.
Demonstrated experience deploying and supporting analytical or machine-learning products in production, beyond experimentation or proof of concept.
Applied experience with multiple machine-learning methods, including regression, classification, clustering, forecasting, optimization, neural networks, natural language processing, generative AI, or related methods.
Experience with Databricks or another modern cloud-based data and machine-learning platform.
Experience with version control, automated testing, CI/CD, model evaluation, deployment, monitoring, and reproducible development practices.
Experience building solutions that use both structured and unstructured enterprise data.
Demonstrated depth in at least two of the following areas: Predictive cost modeling, forecasting, optimization, or decision science; Large language model applications, including retrieval-augmented generation, tool calling, structured outputs, or evaluation frameworks; Document intelligence, information extraction, or supplier quote and cost-breakdown processing; Multimodal machine learning using images, engineering files, or 3D geometry; Scalable data engineering and production ML platforms
Experience building analytical applications, APIs, dashboards, or visualization solutions using tools such as FastAPI, Flask, Dash, Streamlit, Power BI, Tableau, Plotly, or similar frameworks.
Ability to translate business questions into data, product, and modeling requirements.
Demonstrated ability to work effectively across product, engineering, finance, purchasing, and business teams.
Ability to manage multiple priorities and deliver high-quality work in ambiguous situations.
Clear written and verbal communication skills.

Nice To Haves

Master’s degree or higher in a related quantitative discipline.
Experience in automotive engineering, product development, purchasing, supply chain, manufacturing, finance, or cost engineering.
Experience developing cost-estimation, should-cost, profitability, sourcing, supplier-risk, or financial-risk models.
Experience with part-image analysis, computer vision, 3D geometry, CAD-related data, or engineering file processing.
Experience developing Text-to-SQL or natural-language interfaces to enterprise data.
Experience processing supplier quotes, ED&D breakdowns, vendor tooling estimates, technical documents, or other cost-engineering artifacts.
Experience with PySpark, Delta Lake, MLflow, Databricks Workflows, Databricks Asset Bundles, or related platform capabilities.
Demonstrated ability to connect technical delivery to measurable cost, efficiency, adoption, or decision-quality outcomes.

Responsibilities

Partner with Engineering, Cost Engineering, Purchasing, Finance, Program Management, R&D, and business stakeholders to define problems, success measures, product requirements, and delivery priorities.
Develop predictive and statistical models for part-cost estimation, cost-driver analysis, forecasting, classification, optimization, and decision support.
Build and maintain production-grade data pipelines integrating engineering, purchasing, supplier, manufacturing, and financial data.
Develop reusable Python frameworks, automation, and data-processing patterns for complex structured and unstructured data.
Build user-facing analytical applications, APIs, dashboards, and visualizations that turn model outputs into business decisions.
Develop AI capabilities for natural-language access to cost data, supplier quote and document processing, engineering workflow automation, and decision support.
Apply machine learning, large language models, retrieval-augmented generation, tool calling, structured outputs, and AI agents where they create measurable business value.
Develop and evaluate multimodal solutions that may use images, engineering files, 3D geometry, documents, and relational data to support cost-estimation and cost-engineering use cases.
Establish data-quality controls, model-evaluation methods, documentation, and monitoring required for reliable production use.
Use Git, automated testing, CI/CD, MLOps, and model deployment practices to create maintainable, reproducible, and governed solutions.
Communicate technical findings, limitations, recommendations, and business value to technical and non-technical audiences.
Work within GM requirements for data protection, responsible AI, security, governance, and model risk management.
Influence stakeholders through data, technical credibility, and clear product thinking, including in situations with incomplete data or limited precedent.