Data Scientist

Elder Research•Arlington, VA

16d•Hybrid

About The Position

As a Data Scientist, you will support the Internal Revenue Services mission to improve tax compliance, fraud detection, and risk identification across large, complex tax and financial data environments. You will work directly with government clients, program managers, and technical teams to understand business and compliance challenges, design analytical approaches, and deliver data-driven solutions that inform enforcement, audit prioritization, and fraud prevention efforts. In this role, you will develop, test, and deploy predictive and statistical models using structured and unstructured data to identify anomalies, non-compliance risk, and potential fraud within tax records and related datasets. You will contribute across the full data science lifecycle, from problem formulation and data exploration through model validation, deployment, and stakeholder communication.

Requirements

Bachelor of Science degree in a relevant field such as statistics, computer science, economics, mathematics, analytics, data science, business, or social sciences
2-10+ years of experience in data science, analytics, or a related technical field
Experience using version control systems (e.g., Git) and collaborative development practices
Strong understanding of relational databases and SQL
Comfortable learning new tools, methodologies, and domains, including working outside your comfort zone
Strong analytical mindset with a willingness to tackle complex mathematical and statistical challenges
Willingness to travel and work on-site at client locations as required by project needs
Must currently possess an IRS Public Trust clearance with Full Background Investigation
Must be able to remain in a stationary position 50%
Needs to occasionally move about inside the office to access file cabinets, office machinery, etc.
Frequently communicates with co-workers, management, and customers, which may involve delivering presentations.
Must be able to exchange accurate information in these situations

Nice To Haves

Advanced degree (MS or PhD) in statistics, computer science, data science, mathematics, analytics, engineering, or related fields
Experience with PySpark, Unity Catalog, and Jobs in DataBricks
Experience with Natural Language Processing (NLP) and text analytics applied to unstructured documents or case notes
Experience with graph analytics and network analysis to identify relationships, fraud rings, or interconnected entities
Familiarity with platforms and tools such as Databricks, and AWS
Experience with containerization and environment management (e.g., venv, conda)
Understanding of the data analytics lifecycle (e.g., CRISP-DM)
Experience applying advanced statistical concepts including sampling considerations, bias detection, weighting techniques, handling missing or outlier data, exploratory analysis, and longitudinal forecasting
Experience operating in secure or remote government environments, including use of bash and command-line tools

Responsibilities

Prior programming experience, preferably in Python or R, including data exploration, feature engineering, and model development
Explore, clean, and wrangle large, complex datasets to uncover insights and identify opportunities for data sciencedriven solutions in support assessments, gap analyses, and actionable recommendations for IRS stakeholders
Design, develop, test, and implement quantitative and qualitative data science solutions that are modular, maintainable, and adaptable to evolving government and regulatory requirements
Demonstrated experience using Python, SQL, and Databricks for data analysis, modeling, and statistical evaluation
Apply statistical and machine learning techniques to anomaly detection, fraud identification, and non-compliance risk scoring, including supervised and unsupervised approaches to help prioritize cases based on compliance risk, fraud indicators and business impact
Develop and evaluate predictive risk models used to support audit selection, refund review, and fraud prevention initiatives
Test and validate models using robustness, sensitivity, and significance testing, ensuring defensible and explainable results
Write modular, reusable, and well-documented code within an iterative development process that includes peer review and collaboration
Collaborate with clients, subject matter experts, and cross-functional teams to refine problem statements, requirements, and analytical approaches
Prepare and deliver technical and non-technical briefings, reports, and presentations to audiences with varying levels of analytical sophistication
Demonstrated ability to work independently in a collaborative, fast-paced environment
Strong interpersonal, written, and verbal communication skills, with the ability to translate business and compliance needs into technical solutions
Experience using Markdown for technical documentation