Data Scientist – Experimentation & Causal Inference

Design, build, and validate causal models to evaluate the impact of business campaigns and initiatives (2x2 Difference-in-Differences (DiD), Staggered DiD, Synthetic Control, Causal Forest, DoubleML, Causal Meta Learners, G-Computation, Causal Discovery, DAG).

Define causal estimation methodologies for the design, implementation, and validation of causal models, aligned with business context, data quality, and the strategic decisions to be informed, ensuring standards and best practices.

Apply advanced statistical methods on observational data to identify and quantify causal relationships, distinguishing correlation from causation.

Design and implement Randomized Controlled Trials (RCTs) to rigorously evaluate the effectiveness of business strategies.

Ensure robust experimental design, proper allocation of control and treatment groups, and correct execution to obtain reliable and reproducible results. Estimate the causal impact of experiments using appropriate statistical techniques (e.g., regression, t-test, chi-square), evaluating assumptions, biases, and the robustness of results.

Maintain clear and detailed documentation of models, experiments, and analytical processes.

Prepare reports and presentations that translate complex analyses into understandable messages for non-technical audiences.

At least 3–4 years developing and applying causal inference models, experimentation (RCTs), and machine learning techniques on experimental and observational data.

Strong foundations in statistics, probability, and applied econometrics, including the Potential Outcomes framework, selection bias, omitted variable bias, parallel trends, spillovers, and SUTVA.

Experience in advanced experimental design: statistical power analysis, sample size calculation, multiple testing control, and heterogeneity analysis (CATE).

Advanced proficiency in Python and R applied to data analysis, experimentation, and causal inference, with experience in Python libraries such as econml, causalml, dowhy, statsmodels, and scikit-learn, and R packages such as did, fixest, CausalImpact, MatchIt, Synth, and gsynth.

Advanced proficiency in SQL and PySpark for extracting, transforming, and analyzing large datasets in distributed environments.

Experience working in the Azure suite, including Databricks and Azure DevOps, for developing, versioning, and deploying pipelines and analytical workflows.

Understanding of the specific challenges of the Consumer Packaged Goods (CPG) industry.

Ability to operate in fast-paced, highly ambiguous environments.