Data Scientist - ML & Operational Analytics

Job title: Data Scientist - ML & Operational Analytics

Job location: Washington, DC (Hybrid)

Job Type: Contract

Job description:

Role Overview

The Senior Data Scientist – ML & Operational Analytics will sit on the business-facing side of data science, partnering directly with operational and infrastructure stakeholders to define problems, build machine learning solutions, and deploy models into production.

This role is not a backend data engineering or IT support position. It is a full‑lifecycle data science role focused on solving real business problems through predictive modeling, analytics, and AI.

You will support multiple initiatives across Safety and Infrastructure Analytics, with a heavy emphasis on asset health, reliability, efficiency, and operational performance. Approximately 60% of the role is new model development, with the remaining 40% enhancing and maintaining existing models.

Key Responsibilities

Machine Learning & Analytics

• Design, develop, and deploy machine learning models including regression, classification, and time‑series models for operational use cases.
• Apply advanced statistical and ML techniques to large‑scale datasets (terabytes to petabytes), including:
• Smart‑meter data
• Smart‑grid and IoT data
• Structured (relational databases)
• Unstructured data (text, documents, and limited multimedia)
• Perform feature engineering, data validation, and quality assessment to ensure model reliability and interpretability.
• Enhance existing models and pipelines while leading the development of net‑new solutions.

Business Partnership & Problem Solving

• Work directly with business stakeholders to:
• Identify operational problems
• Translate business needs into analytical frameworks
• Define success metrics and model outcomes
• Clearly communicate analytical findings, model results, and recommendations to non‑technical audiences.
• Validate insights with the business and iterate based on feedback.
• Own solutions end‑to‑end: problem → data → model → deployment → business adoption.

Data Science Lifecycle & Collaboration

• Collect, cleanse, standardize, and analyze data from multiple internal and external sources.
• Collaborate closely with:
• Information architects
• Data engineers
• Project and program managers
• Other data scientists and analysts
• Ensure smooth handoff and adoption of deployed solutions.
• Document methodologies, assumptions, and results to support governance and reuse.
• Act as a subject matter expert in machine learning, AI, feature engineering, data mining, and statistical modeling.

Required Qualifications

• MS degree in Computer Science, Statistics, Mathematics, Engineering, Physics, or a related quantitative field
  (or 15+ years of equivalent professional data science experience)
• 5+ years of hands‑on experience as a data scientist working on operational analytics or applied ML problems.
• Proven experience building and deploying ML models—not just training or research models.
• Strong proficiency in:
• Python (primary)
• R
• SQL
• Common ML libraries (e.g., scikit‑learn, stats models, etc.)
• Strong foundation in:
• Probability and statistical inference
• Regression techniques
• Experimental design and validation
• Demonstrated experience working closely with business stakeholders to deliver production solutions.

Preferred Qualifications

• PhD in Computer Science, Statistics, Mathematics, Engineering, Physics, or related field.
• Experience within an Electric Utility, Energy, Infrastructure, or Industrial environment.
• Hands‑on experience with Azure Machine Learning for model development and deployment.
• Knowledge of optimization techniques, including:
• Linear programming
• Mixed‑integer optimization
• Exposure to:
• Computer vision
• Generative AI use cases
• Azure certifications are a plus.

Must Have Candidate Criteria

• Must have owned data science solutions end to end, including defining the business problem and KPIs, sourcing and validating data, building and deploying ML models, and supporting the solution after deployment in an operational environment.
• Must have hands on experience building and deploying machine learning models for real operational use cases, specifically regression, classification, and/or time series models, with demonstrated feature engineering and appropriate model evaluation techniques.
• Must be proficient in SQL, Python, R, and data preparation for modeling, including the ability to independently create training datasets, perform data quality checks, prevent data leakage, and work with large, complex datasets without relying on a separate data engineering team.
• Must have a strong statistical foundation and validation mindset, including experience with inference, experimental or quasi experimental design, regression analysis, and clearly proving model impact beyond accuracy metrics.
• Must have demonstrable experience partnering directly with business stakeholders, translating ambiguous business problems into analytical solutions, communicating findings to non technical audiences, and influencing decisions based on data driven insights.