Senior Data Scientist (NLP, Unstructured Data)

Salary: $70,000 - 110,000 per year Requirements:

• We require a masters, Ph.D., doctorate-level equivalent, or comparable background in data science, machine learning, computer science, mathematics, or a related discipline; alternatively, we accept 10 years of applied experience in one of these fields.
• We need 5+ years of experience designing, implementing, and sustaining advanced AI and predictive systems, including supervised and unsupervised methods.
• We look for 5+ years of experience developing analytic rules and models with modern analytical tools and proven best practices.
• We require 5+ years of experience building regression, classification, and other statistical models to surface anomalies, patterns, and predictive factors.
• We need 3+ years of experience supporting criminal investigations tied to financial fraud or misuse of government funds.
• We require 3+ years of Python data manipulation experience, with Pandas as essential.
• We look for 3+ years of experience working in a modern cloud environment such as Azure, AWS, or Google Cloud Platform; certifications are preferred.
• We require 2+ years of advanced SQL analysis experience, specifically with SQL Server and PostgreSQL.
• We need 2+ years of experience developing and scaling natural language processing solutions.
• We look for 2+ years of experience presenting methods and findings to both technical and non-technical audiences through verbal, written, and visual formats.
• We prefer candidates with production experience in named entity recognition and entity resolution across challenging document collections.
• We prefer experience with retrieval-augmented generation, vector stores, embeddings, and large-scale semantic search.
• We prefer experience integrating large language models under federal security constraints, including boundary-controlled deployment and prompt versioning.
• We prefer experience building optical character recognition workflows for scanned or low-quality source documents.
• We prefer experience with topic modeling, document classification, or clustering in audit, legal, or investigative text.
• We prefer cloud certification in Azure, AWS, or Google Cloud Platform.

• We integrate and scale natural language processing techniques to parse, clean, and analyze large volumes of unstructured and semi-structured text, using OCR, semantic similarity methods, and large language models when needed.
• We design, develop, test, calibrate, and deploy statistical and machine learning models focused on financial fraud, improper payments, and noncompliance within SBA programs.
• We build and refine supervised and unsupervised approaches, including regression, Bayesian, clustering, and ensemble techniques.
• We review, maintain, and support existing loan fraud indicators developed by TSD.
• We conduct data quality assessments on source tables and create repeatable workflows for combining and analyzing large data sets.
• We partner directly with criminal investigators to define and execute analytic strategies for loan fraud cases while following federal criminal procedure rules governing protected information, including Rule 6(e).
• We generate case leads for SBA OIG investigations from model outputs.
• We document methodology, test models, and production models in a format that meets criminal evidentiary standards.
• We create visualizations and dashboards that communicate methodological decisions, outcomes, and predictive performance, refining them based on end-user feedback.
• We present findings in multiple formats, including data summaries and visualizations for investigative teams and executive-level summaries for OIG leadership.
• We coordinate with the data engineering function to ensure the architecture efficiently supports machine learning and text-processing pipelines.
• We develop automation and programming solutions using SharePoint, Python, Excel, Power BI, Power Apps, and related tools.
• We identify new business questions that broaden the scope of analysis and reporting.

• AI
• AWS
• Azure
• Cloud
• Excel
• Support
• Machine Learning
• PostgreSQL
• Power BI
• Python
• SQL
• Security
• SharePoint
• pandas

More:

We are hiring a Senior Data Scientist for a remote role based in Herndon, VA. This position requires a Public Trust Clearance and offers the opportunity to work on NLP and unstructured data analytics in support of SBA fraud, improper payment, and noncompliance investigations. We provide a competitive compensation and benefits package that includes medical, dental, vision, basic life, a health savings account, 401(k) matching, three weeks of PTO/sick time, 11 paid holidays, and pre-approved online training.

last updated 33 week of 2026