Data Scientist 1

About Us

Here at Baylor Scott & White Health we promote the well-being of all individuals, families, and communities. Baylor Scott and White is the largest not-for-profit healthcare system in Texas that empowers you to live well.

Our Core Values are:

• We serve faithfully by doing what's right with a joyful heart.
• We never settle by constantly striving for better.
• We are in it together by supporting one another and those we serve.
• We make an impact by taking initiative and delivering exceptional experience.

Benefits

Our benefits are designed to help you live well no matter where you are on your journey. For full details on coverage and eligibility, visit the Baylor Scott & White Benefits Hub to explore our offerings, which may include:

• Immediate eligibility for health and welfare benefits
• 401(k) savings plan with dollar-for-dollar match up to 5%
• Tuition Reimbursement
• PTO accrual beginning Day 1
  

Note: Benefits may vary based on position type and level.

Work Authorization Requirement: Candidates must be legally authorized to work in the United States
without employer sponsorship. Visa sponsorship is not available for this position at this time.

About the Team

Care Coverage Analytics supports the data science and advanced analytics needs of Baylor Scott & White
Health and the Baylor Scott & White Quality Alliance (ACO). We support over 1.2 million lives across multiple
blocks of business by partnering with operational teams and subject-matter experts to improve healthcare
decisions and outcomes.

Job Summary
We are seeking a Data Scientist 1 to develop practical data science, natural language query (NLQ), and agentic
AI solutions for complex healthcare problems. Serving as both a data scientist and a consultative partner, you
will work with data scientists, data engineers, subject-matter experts, and business partners to frame
problems, challenge assumptions, and turn questions into reliable models and AI-enabled solutions that
improve access to trusted information, reduce manual work, and support better decisions.
In this role, you will contribute across the solution lifecycle, from problem definition and data preparation
through development, evaluation, deployment, and monitoring. You will study how people use NLQ and
agentic products and help turn behavioral patterns, feedback, and unmet needs into product improvements.

You will also use AI-assisted coding tools and learn to package repeatable work as reusable agent skills and
internal AI solutions. This applied role requires curiosity, sound data science foundations, and a commitment
to accuracy, security, explainability, and responsible AI.

Responsibilities and Time Allocation

Communication and Consulting | 25%: Build trusted relationships with business, clinical, and technical
partners. Clarify needs, frame problems, challenge assumptions, recommend appropriate analytical
approaches, establish success measures, and communicate findings, tradeoffs, and limitations in language that
supports decisions and action.

NLQ and Agentic AI | 20%: Develop and improve grounded NLQ applications, data science copilots, reusable
agent skills, and multi-step AI workflows for healthcare questions and recurring technical work. Contribute to
internal tools that accelerate development and test whether outputs are accurate, useful, and reproducible.

Applied Data Science and Experimentation | 20%: Apply statistical analysis, machine learning, natural
language processing, behavioral segmentation, and other appropriate techniques to healthcare and product
questions. Establish baselines, analyze user journeys and cohorts, design experiments, compare solution
options, document assumptions, and communicate limitations.

Model and Agent Lifecycle | 15%: Support models and AI agents from experimentation through deployment
and monitoring. Use validation checks, interaction patterns, feedback, task outcomes, and error analysis to
monitor adoption and performance drift and improve reliability, explainability, security, and user experience.

Data Preparation and Feature Development | 10%: Use SQL and Python to explore, prepare, and validate
claims, EHR, member, quality, and operational data. Develop reproducible modeling datasets and features,
assess data quality and bias, and use governed metrics, semantic models, metadata, and business definitions
to ground NLQ and agentic solutions in consistent information.

Continuous Learning and Knowledge Sharing | 10%: Participate in peer reviews, document methods, seek
feedback, share knowledge with teammates, and stay current with developments in data science, generative
AI, agentic systems, healthcare AI, and responsible AI.

Primary Responsibilities

• Consult with business, clinical, and technical partners to clarify needs, frame healthcare problems,
  challenge assumptions, and define testable hypotheses and measurable success criteria.
• Use SQL and Python to explore and prepare data, engineer features, perform statistical analysis,
  develop models, automate workflows, and validate results.
• Develop grounded NLQ applications, data science copilots, and agentic workflows using appropriate
  machine learning, natural language processing, retrieval, and tool-use techniques.
• Evaluate models and agents for accuracy, task completion, consistency, source quality, latency, cost,
  safety, and user value; investigate failures and recommend improvements.
• Analyze privacy-approved interaction data to identify behavioral segments, unmet needs, and recurring
  failure patterns, then translate those findings into evaluation cases and product hypotheses.
• Support experiments, pilots, and cohort analyses that measure whether product changes improve task
  success, adoption, trust, and business or clinical outcomes.
• Apply responsible AI, security, data-minimization, de-identification, access, and retention practices to
  healthcare data and product telemetry.
• Use approved AI coding assistants responsibly and, under guidance, help turn repeatable work into
  reusable prompts, agent skills, templates, or internal AI workflows for data science and analytics teams.
• Develop reusable modeling datasets, features, metrics, tests, documentation, and monitoring controls
  while participating in source control, code review, deployment, and production support.
• Present findings, assumptions, limitations, options, and recommendations clearly so technical and non-technical partners can make informed decisions and take action.

Strongly Preferred Qualifications
We value demonstrated ability, continuous learning, and collaborative problem-solving. Candidates are not
expected to bring every qualification.

• Proficiency in Python and SQL for data exploration, feature engineering, statistical analysis, modeling,
  and automation.
• Academic or professional experience with statistical analysis, experimentation, classification, regression,
  natural language processing, or related machine learning methods, including evaluation through
  quantitative metrics and error analysis.
• Strong problem-solving and communication skills, including the ability to investigate unexpected
  results and explain methods, assumptions, limitations, and recommendations to non-technical partners.
• Experience working collaboratively through source control, code review, documentation, and testing.
• Familiarity with AI coding assistants and the judgment to review, test, and refine AI-generated code or
  technical recommendations.

Ideal Candidate Demonstrates

• Experience using NLQ applications, AI assistants, or coding copilots and understanding concepts such
  as prompting, grounding, retrieval, tool use, structured outputs, and hallucination risk; interest in
  learning to build reusable agent skills is important.
• Experience evaluating generative AI or analyzing product telemetry, user journeys, cohorts, behavioral
  segments, and experiments to improve a digital or AI-enabled product.
• Experience with feature engineering, experiment design, model selection, model validation, or
  interpretable machine learning; familiarity with semantic models and governed metrics is beneficial for
  NLQ work.
• Familiarity with healthcare data and its secure use, including claims, electronic health records, member
  data, quality measures, data classification, masking, role-based access, and protected health
  information.
• Experience with Snowflake or another cloud data platform, including deployment, monitoring,
  automated testing, CI/CD, or production support for data science and machine learning solutions.
• Curiosity about how people ask questions and make decisions, coupled with sound judgment about
  when generative AI, machine learning, statistical analysis, or a simpler solution is the best fit.

Belonging Statement
We believe that all people should feel welcomed, valued, and supported.

QUALIFICATIONS

• EDUCATION - Masters' or Bachelors plus 2 years of work experience above the minimum qualification
• EXPERIENCE - 3 Years of Experience