Data Scientist

Headwater Science; Durham (NC); On-site; USD 90,000 - 150,000; Full time

Benefits offered by this job

Hybrid work

Health coverage for you and family

401(k) with company match

Generous PTO and holidays

Paid parental leave

Job summary

Headwater Science in Durham, NC is seeking a Data Scientist to translate research protocols into reproducible analyses. You will write R code to construct cohorts, run analyses, and produce study results as tables and figures.

You will collaborate with cross-functional teams, manage pipelines in GitLab, and ensure the analytic datasets and models are well-documented and reproducible for client deliverables. Nice-to-have includes experience with healthcare data.

Qualifications

• Master’s degree or 2+ years in biostatistics, epidemiology, health economics, bioinformatics, data science, or related field.
• Strong R programming skills.
• Experience querying relational databases via SQL or dbplyr.
• Ability to write readable, reproducible code.
• Strong communication and collaboration with cross-functional teams.

Responsibilities

• Develop R code to execute research projects, including cohort construction and analyses that generate study results.
• Coordinate development through GitLab as code is assembled into project deliverables.
• Build analytic datasets with baseline characteristics, time-varying treatments, outcomes, and censoring.
• Fit models per study design and run diagnostics like propensity score distributions and covariate balance.
• Create study results components and client deliverables (tables/figures).
• Extend the project pipeline from analytic-ready datasets through finished reports.

Skills

Strong R programming

Strong communication

Readable, reproducible code

Education

Master’s degree

2+ years relevant work experience

Biostatistics / epidemiology background

Tools

dbplyr

SQL

GitLab

causalRisk

Job description

Headwater Science partners with clients to advance healthcare research by converting large, messy source data into analysis-ready datasets and producing reproducible statistical outputs. This hybrid role in Durham, NC supports research studies end to end, from cohort construction and eligibility logic through modeling, diagnostics, and report-ready tables and figures.

As a Data Scientist, you will write R code to implement research protocols and SAP specifications, using quality control practices and documentation to ensure results can be reproduced each time a pipeline runs.

What you will do

• Develop R code to execute research projects, including cohort construction and running analyses that generate study results.
• Coordinate development through GitLab as code is assembled into project deliverables.
• Support the cohort building and analysis/reporting phases of research projects, working from study protocols and study design decisions.
• Apply domain knowledge of how clinical events are represented across different databases to choose appropriate study variables.
• Use dbplyr to query relational source databases from R, generating SQL for tables that are too large to load fully into memory.
• Implement eligibility criteria from the study protocol and/or SAP to identify the study population.
• Determine the index date and start of follow-up for each patient.
• Build analytic datasets, often organized as one row per patient with baseline characteristics, time-varying treatment and health status changes, outcomes of interest, and censoring.
• Fit models specified in the study design, including use of causalRisk, Headwater Science’s internal causal inference package for observational data.
• Run analysis diagnostics such as checking propensity score distributions, weight behavior, and covariate balance, and flag issues back to the study design team.
• Create study results components including tables and figures such as baseline characteristics, effect estimates, risk curves, and treatment patterns, then assemble them into client deliverables.
• Build project repositories as orchestrated pipelines so study results can be reproduced reliably whenever the pipeline is executed.

Quality control and collaboration

• Participate in quality control procedures including code review, testing, and double coding of analysis portions as an independent check.
• Work closely with the study design team across the study lifecycle, incorporating changing requirements, sharing interim results, and providing feedback on how well the study design aligns with the available data.
• Extend the project pipeline from the analytic-ready dataset through to the finished report.

What you bring

• Master’s degree, or 2+ years of relevant work experience in biostatistics, epidemiology, health economics, bioinformatics, data science, or a related quantitative field.
• Strong R programming skills.
• Experience querying relational databases, either directly in SQL or via tools like dbplyr.
• Experience writing code that others can read, run, and reproduce.
• Strong communication, presentation, and collaboration skills with cross-functional technical and scientific teams.

Tools and technologies

• R
• GitLab
• dbplyr
• SQL
• causalRisk

Benefits

• Hybrid work
• Comprehensive health, dental, and vision coverage for you and your family
• 401(k) with company match
• Generous PTO and company holidays
• Paid parental leave

Nice to have

• Experience with real-world healthcare data such as administrative claims, EHR, hospital chargemaster, or registry data
• Experience constructing longitudinal datasets for observational research