Talent Groups

Talent Groups

Data Scientist A/B testing

Austin, Texas, USA · full-time

Sponsorship not specified$120k-$125kDetected 63 days ago
PythonSQLData ScienceData VisualizationStatisticsA/B TestingGoogle AnalyticsExperimental DesignLeadershipCritical Thinking

About the role

  • Job Description: A/B Test Data Scientist (Website Experimentation) Role Overview As an A/B Test Data Scientist, you will be the statistical engine behind our website’s optimization strategy.
  • You will work at the intersection of product, engineering, and marketing to ensure every website change is backed by rigorous data.
  • Determine appropriate sample sizes, statistical power, and experiment duration for reliable test outcomes.

Responsibilities

  • Develop clear experimentation hypotheses and define primary and secondary success metrics.
  • Manage experiment execution, including user segmentation and randomization processes.
  • Collaborate with engineering teams to ensure accurate event tracking and data integrity.
  • Maintain and manage an experimentation roadmap while documenting and archiving key learnings and outcomes.
  • You will design, execute, and analyze controlled experiments to improve user experience, conversion rates, and business growth.

Requirements

  • Knowledge of Bayesian vs.
  • Proficiency in SQL for data extraction and Python or R for advanced statistical analysis.
  • Required Skills & Qualifications Statistical Expertise: Mastery of hypothesis testing, experimental design, and probability.
  • Programming: Proficiency in SQL for data extraction and Python or R for advanced statistical analysis.

Nice to have

  • Bachelor’s degree in a quantitative field (Statistics, Math, CS, Economics)
  • master’s or PhD preferred.
  • Frequentist approaches is a plus.

Skills

  • Experience with experimentation platforms (e.g., Optimizely, VWO, or LaunchDarkly) and web analytics (e.g., Google Analytics 4).

Compensation

  • $120k-$125k

This listing is sourced directly from Talent Groups's careers page and normalized into a canonical job model.