Director Data Science, Measurement
@ SambaDirector Data Science, Measurement
About the job
Samba is a media intelligence company providing real-time consumer insights at scale. The Director leads measurement science, building models for attribution, incrementality, and audience targeting, working closely with product and engineering teams.
Requirements
- 8+ years data science experience
- Expertise in Causal ML
- Strong statistical knowledge
- Python and SQL proficiency
- Experience leading teams
Qualifications
- Bachelor's in statistics or related
- Master's or PhD preferred
- Experience with measurement tools
- Strong communication skills
- Background in advertising measurement
Full job description
Reporting to our VP of Data Science, Samba is looking for a Director of Data Science to lead our Measurement science team - the group responsible for building the statistical frameworks, causal models, and attribution methodologies that power Samba's core measurement products. You will own the science and delivery for a portfolio spanning incrementality measurement, multi-touch attribution, reach and frequency modeling, and audience intelligence, working closely with Product to shape direction and with Engineering to bring solutions to production.
You are a senior technical leader first. You bring deep, hands-on expertise in causal ML, statistical modeling, and measurement science - enough to drive architectural decisions, mentor senior data scientists, and engage credibly in design reviews. You also know how to build and run a high-performing team, communicate clearly to executive and external audiences, and keep complex multi-workstream delivery on track.
WHAT YOU'LL DO
Measurement Science Strategy: Partner with Product to define the measurement science roadmap - spanning incrementality, multi-touch attribution, reach/frequency estimation, panel calibration, and audience targeting. Translate business and client priorities into well-scoped quarterly plans and sprint commitments.
Technical Leadership: Lead design reviews and architecture decisions across the team. Drive the application of Causal ML - counterfactual modeling, meta-learners, and heterogeneous treatment effect estimation - as the primary framework for measuring ad effectiveness, alongside rigorous command of the broader toolkit: A/B testing, DiD, synthetic control, Bayesian hierarchical models, and panel methodology.
Delivery & Execution: Own end-to-end delivery across the team's measurement science portfolio - managing dependencies, removing blockers, and ensuring timely delivery of high-quality work across multiple concurrent workstreams.
Scientific Standards & Data Quality: In partnership with Engineering, set and maintain standards for experimental design, model evaluation, reproducibility, and production readiness. Co-own the data quality framework that ensures the robustness and consistency of measurement products used by clients and internal stakeholders.
Best Practices & MLOps: In partnership with Engineering, develop and implement best practices for the full DS lifecycle - data management, modeling, evaluation, pipeline orchestration, and production deployment on Databricks/Spark.
People & Team Leadership: Lead, mentor, and grow a team of data scientists - owning hiring, performance reviews, career development, and individual goal-setting aligned to Samba's leveling framework. Build a high-accountability culture grounded in technical rigor, psychological safety, and continuous learning.
Cross-functional Partnership: Collaborate with Product, Engineering, and Business Development to define and execute data-driven measurement projects. Support Sales and Marketing with technical positioning and clear articulation of Samba's measurement capabilities.
Stakeholder Communication: Translate complex causal and statistical findings into clear, actionable insights for technical and non-technical audiences. Represent the team to senior leadership and external stakeholders, and contribute a credible technical voice to client conversations and industry forums.
WHO YOU ARE
8+ years of hands-on data science experience with at least 2-3 years in a people management role, including demonstrated ability to hire, develop, and retain senior data scientists
Deep, first-principles expertise in Causal ML - counterfactual modeling, meta-learners, and heterogeneous treatment effect estimation - applied to advertising measurement or media outcomes; familiarity with EconML, DoWhy, or CausalML a plus
Solid command of the broader measurement science toolkit: A/B testing, difference-in-differences, synthetic control, propensity scoring, Bayesian hierarchical models, and panel methodology - with clear intuition for when to apply each approach and what its limitations are
Strong statistical and ML foundations - regression, classification, experimental design, model evaluation, and the ability to reason clearly about trade-offs between modeling approaches
Expert-level Python and SQL; strong PySpark and Databricks experience for large-scale measurement pipelines
Track record of owning and delivering complex, multi-workstream data science projects on time in a fast-moving environment
Excellent communicator - able to translate causal and statistical reasoning into language that drives product, sales, and executive decisions
Bachelor's degree required in Statistics, Computer Science, Mathematics, or a related quantitative field; Master's or PhD strongly preferred
Direct experience with TV or digital measurement - ACR/STB data, viewership panels, reach/frequency modeling, or cross-platform measurement (linear + CTV/OTT)
Hands-on experience with multi-touch attribution (MTA) or multi-channel attribution modeling - understanding of rule-based limitations and the methodological trade-offs of data-driven alternatives
Familiarity with the measurement vendor landscape (Nielsen, Comscore, VideoAmp, iSpot) and industry standards (MRC accreditation, GRP/TRP frameworks)
Experience with audience segmentation, identity resolution, or privacy-preserving measurement approaches
Track record of publishing research, white papers, or presenting at industry conferences