Senior Data Scientist

About the role

Node.Digital LLC is seeking a specialist to build predictive models and natural language processing pipelines to detect financial fraud within Small Business Administration programs. You will work alongside criminal investigators to turn data patterns into actionable leads while ensuring all work meets strict evidentiary standards.

Key facts

What you'll do

• Apply natural language processing, optical character recognition, and semantic similarity to analyze unstructured text.
• Create and maintain statistical and machine learning models to identify improper payments and non-compliance.
• Build supervised and unsupervised models using regression, Bayesian, clustering, and ensemble methods.
• Manage existing loan fraud indicators for the TSD team.
• Partner with criminal investigators to develop analytic strategies that comply with federal rules of criminal procedure, including Rule 6(e).
• Generate case leads for SBA OIG investigations.
• Document methodologies to satisfy criminal evidentiary requirements.
• Create dashboards and summaries for both investigative staff and OIG leadership.
• Coordinate with data engineering to optimize machine learning and text processing pipelines.
• Automate workflows using Python, SharePoint, Power BI, Power Apps, and Excel.

Requirements

• Master degree, PhD, or equivalent in data science, mathematics, computer science, or machine learning, or 10 years of relevant work experience.
• 5+ years of experience building and maintaining AI systems and predictive models.
• 5+ years of experience using analytic tools to develop models and rules.
• 5+ years of experience with regression, classification, and statistical modeling for pattern detection.
• 3+ years of experience supporting criminal investigations regarding financial fraud or government fund abuse.
• 3+ years of experience using Python with Pandas.
• 3+ years of experience working in Azure, AWS, or GCP cloud environments.
• 2+ years of experience performing data analysis in SQL Server and PostgreSQL.
• 2+ years of experience scaling natural language processing solutions.
• 2+ years of experience presenting technical findings to diverse audiences.
• Must hold a Public Trust Clearance.

Nice to have

• Experience with named entity recognition and entity resolution in large document sets.
• Proficiency in retrieval augmented generation, vector stores, embeddings, and semantic search.
• Background in deploying large language models within federal security constraints.
• Experience with optical character recognition on low-quality documents.
• Skills in topic modeling, document classification, or clustering for legal or audit text.
• Cloud certifications in Azure, AWS, or GCP.

Skills & tools

• Python, Pandas, SQL Server, PostgreSQL, Power BI, Power Apps, SharePoint, Excel, Azure, AWS, GCP, NLP, Machine Learning.

Practical notes

• Benefits include medical, dental, vision, basic life insurance, health savings account, 401k matching, 3 weeks of PTO/sick leave, 11 paid holidays, and access to pre-approved online training.