Data Scientist Leader — Experimentation & Causal Inference

Unchain Data; San Francisco (CA); On-site; USD 170,000 - 260,000; Full time

Job summary

Unchain Data is seeking a Senior Data Scientist specializing in experimentation and causal inference to own the statistical framework atop our platform. You will design, validate, and improve methods to ensure trustworthy, sensitive, and correctly interpreted experiments.

You will collaborate with a team of data scientists and platform engineers in the Big Data group, implementing guardrail metrics, variance reduction techniques, and continuous A/A monitoring, and you will translate complex

Qualifications

• MS or PhD in Statistics, Biostatistics, Econometrics, CS or related field.
• 3+ years designing and analyzing online controlled experiments at tech scale.
• Strong foundations in hypothesis testing, power analysis, multiple testing, sequential testing.

Responsibilities

• Design and maintain automated anomaly detection for live experiments, SRM checks and traffic validation.
• Define guardrail metrics, thresholds, and circuit-breaking criteria for trustworthy experiments.
• Implement CUPED and pre-experiment covariate adjustment to reduce variance.
• Develop methods to isolate true treatment effects in low-traffic or high-priority cases.
• Collaborate with product and engineering on experiment design, sizing, and interpretation; translate methods into production specs.

Skills

Python

SQL

Experiment design

Statistical analysis

Communication

Education

MS/PhD in Statistics, Biostatistics, Econometrics, CS

Tools

Claude Code

OpenClaw

AI tooling

Job description

Unchain Data is seeking a Senior Data Scientist specializing in experimentation and causal inference to own the statistical framework atop our platform. You will design, validate, and improve methods to ensure trustworthy, sensitive, and correctly interpreted experiments.

You will collaborate with a team of data scientists and platform engineers in the Big Data group, implementing guardrail metrics, variance reduction techniques, and continuous A/A monitoring, and you will translate complex