Natural Language Processing Data Scientist

We are seeking a Natural Language Processing (NLP) Data Scientist with expertise in (1) information extraction using advanced NLP methodologies and (2) natural language data curation to help design and lead rigorous research that informs real-world healthcare decisions. In this role, you will conceptualize research questions, build and maintain robust natural language datasets and NLP pipelines, and develop innovative approaches to determine how healthcare interventions affect patient outcomes. You will apply NLP methodologies to extract complex clinical data from unstructured text sources for incorporation into extensive clinical datasets used for research that can guide clinical practice, policy decisions, and improvements in healthcare delivery for patients.

Working with one of the largest integrated medical records databases in the United States-containing comprehensive clinical information on more than 22 million patients-you will collaborate with multidisciplinary teams of clinicians, statisticians, NLP researchers, and data scientists to investigate pressing healthcare questions. You will identify key variables to extract from medical texts, design annotation frameworks, and lead dataset creation efforts for the development of reliable, accurate ground truth data, and will develop and maintain corresponding pipelines using a range of NLP algorithms to reliably extract these variables and transform them into structured data.

This role offers the opportunity to design impactful studies, develop rigorous pipelines, and translate complex data into insights that influence healthcare policy, clinical practice, and patient outcomes. Your work will contribute directly to evidence-based improvements in care for patients while advancing the scientific application of NLP in large-scale healthcare data.

PRIMARY DUTIES:

• Lead the planning, design, and implementation of advanced information extraction and analysis techniques within a healthcare domain, applying NLP approaches in Python.
• Direct the creation, annotation, and management of high-quality, validated natural language datasets using tools such as Label Studio, maintaining data quality and integrity throughout the project lifecycle.
• Develop, maintain, and optimize robust data and algorithm pipelines, overseeing the end-to-end management of data processes in accordance with data science lifecycle best practices.
• Contribute to continual improvement of the group’s NLP and data science infrastructure, creating reusable codebases and resources for interdisciplinary use across studies and projects involving researchers, engineers, and clinicians.
• Collaborate with clinicians, data scientists, and other stakeholders to guide and execute research initiatives, providing technical input as a subject matter expert to leverage NLP methodologies to address complex clinical requirements.
• Effectively prioritize and manage multiple concurrent projects, tracking progress and ensuring timely completion of project milestones and deliverables.
• Communicate analytical findings, project results, and insights to diverse stakeholders through written reports, presentations, and scientific manuscripts tailored to technical and non-technical audiences.
• Stay current with advancements in NLP, machine learning, and data science, implementing best practices to continuously improve the quality, efficacy, and impact of research and analytic activities.

• MUST be a US Citizen and MUST clear a US government background check

• Degree: Mathematics, statistics, computer science, data science or field directly related to the position. The degree must be in a major field of study (at least at the baccalaureate level) that is appropriate for the position.

• OR -
• Combination of education and experience: Courses equivalent to a major field of study (30 semester hours) as shown above, plus additional education or appropriate experience., * Experience with locally hosted open-source LLMs (such as GPT-OSS, Llama, and Mixtral) and LLM-based pipelines in python for information extraction.
• Experience with natural language dataset creation through expert annotation and data pipeline development/maintenance.
• Experience with SQL/relational databases.
• Experience with Git/Github.
• Previous experience with healthcare data and electronic medical records is highly preferred.
• Strong verbal and written communication skills. Ability to convey complex statistical concepts to non-statistical audiences effectively.
• Experience reading and writing scientific manuscripts.
• Ability to work effectively in multidisciplinary teams and manage multiple collaborative research projects.
• Strong analytical skills and the ability to solve complex problems independently.
• Passion for advancing healthcare research through innovative use of data.

Pulled from the full job description

• Health insurance
• Retirement plan
• Paid time off
• Vision insurance
• Health savings account
• Dental insurance
• Flexible spending account, * Dental insurance
• Employee assistance program
• Flexible spending account
• Health insurance
• Health savings account
• Life insurance
• Paid time off
• Retirement plan
• Vision insurance

This position is based within VA Boston Healthcare System, a leading healthcare system in New England that serves as a premier referral center and a major recipient of research funding. The informatics group is dedicated to leveraging data science and clinical data to conduct high-impact research aimed at expanding our understanding of human health and improving healthcare outcomes.