Technical Analytics Manager / Lead Data Scientist

Hourly pay range: $72.50–$74.86
Pricing level: Level III
Employment type: Full-time
Schedule: Approximately 1,920 hours annually
Location: Primarily remote, with occasional onsite work in Washington, DC

Position Summary

The Technical Analytics Manager / Lead Data Scientist will provide technical and managerial leadership for the development, testing, deployment, maintenance, and quality control of fraud-detection rules, statistical models, machine-learning solutions, risk indicators, anomaly-detection methods, and investigative analytics.

The individual will manage technical data-science activities and ensure that analytic methods, source code, models, findings, visualizations, and supporting documentation meet government standards for accuracy, reliability, reproducibility, explainability, and investigative usefulness.

Primary Responsibilities

• Lead the design and development of analytic rules and models used to identify fraud, waste, abuse, and mismanagement.
• Manage technical assignments, data-science activities, schedules, priorities, dependencies, risks, and quality-control reviews.
• Identify and prioritize innovative fraud-detection use cases based on program risks, available data, investigative needs, and expected mission value.
• Translate fraud typologies, investigative questions, and program risks into actionable analytic approaches.
• Design analytical solutions using statistical analysis, machine learning, anomaly detection, predictive modeling, risk scoring, natural-language processing, graph analytics, and related techniques.
• Direct model development, feature engineering, testing, validation, documentation, deployment, and maintenance.
• Conduct technical and quality reviews of source code, analytic logic, models, findings, visualizations, and technical documentation.
• Develop and maintain analytic rules and models using Python and other open-source programming languages or tools.
• Establish coding standards, model-development procedures, peer-review requirements, testing criteria, version-control practices, and documentation standards.
• Monitor model performance, false-positive rates, false-negative risks, drift, data-quality effects, and changes in fraud patterns.
• Lead model recalibration, redevelopment, replacement, or retirement when source data, program rules, operational needs, or fraud risks change.
• Evaluate the reliability, explainability, scalability, and operational usefulness of analytic models.
• Coordinate with data engineers to ensure that required data is properly ingested, transformed, validated, and made available.
• Coordinate with graph data scientists to incorporate relationship-based and network-based risk indicators.
• Collaborate with investigative analysts and forensic accountants to validate findings and refine analytic leads.
• Prepare technical presentations, model documentation, methodology descriptions, findings, limitations, and recommendations.
• Brief technical and executive government audiences on analytic approaches, results, risks, and recommended actions.
• Support transition and continued operation of existing production models without disrupting government operations.
• Track technical progress and identify delivery, quality, data, model, and operational risks.
• Support the development and execution of quality-control plans for analytics and investigative products.

Required Qualifications

• Minimum of five years of hands-on experience developing analytic rules and models for fraud-detection use cases.
• Minimum of five years of experience designing analytic approaches and managing model-development and model-testing activities.
• Minimum of five years of experience conducting quality control of analytic models, source code, findings, and technical work products.
• Minimum of five years of hands-on experience coding analytic rules or models using open-source programming languages and tools.
• Minimum of five years of experience tracking technical progress and identifying and mitigating delivery risks.
• Demonstrated experience developing innovative use cases addressing fraud, waste, abuse, and mismanagement.
• Strong knowledge of statistical modeling, machine learning, model validation, feature engineering, risk scoring, and anomaly detection.
• Strong Python programming skills.
• Experience with commonly used data-science, statistical, and machine-learning libraries.
• Experience developing, testing, deploying, and maintaining production analytics.
• Demonstrated ability to evaluate model quality, performance, explainability, and operational effectiveness.
• Experience conducting code reviews, model reviews, peer reviews, and technical quality-control activities.
• Experience managing or coordinating multidisciplinary analytics teams.
• Ability to communicate complex technical concepts to investigative, management, and executive audiences.
• Excellent written, verbal, presentation, management, and technical-leadership skills.
• Ability to complete federal suitability, HSPD-12/PIV credentialing, and system-access requirements.

Preferred Qualifications

• Experience with Azure Databricks, MLflow, Microsoft SQL Server, Power BI, Neo4j, Apache Spark, Unity Catalog, or comparable technologies.
• Experience supporting federal-benefit programs, federal oversight organizations, Inspectors General, federal law enforcement, or emergency-relief programs.
• Experience with entity resolution, graph analytics, natural-language processing, large language models, or robotic process automation.
• Experience developing fraud models using large, complex, multi-source government datasets.
• Experience integrating analytic models into operational investigative workflows.
• Graduate degree in data science, statistics, mathematics, computer science, engineering, operations research, economics, or a related discipline.