Principal Data Scientist - Causal Inference / ML

Job Description

A leading transaction data intelligence firm powering spend analytics, marketing measurement, and predictive modeling for Fortune 500 clients is seeking a Principal Data Scientist - Quantitative Intelligence to build the downstream statistical and predictive intelligence that powers their products. You’ll ship end-to-end ML systems, develop rigorous statistical methodologies, and build predictive models on large scale consumer transaction data. This role is hands on, deeply technical, and central to how we transform cleaned and resolved spend data into high value insights.

\n

\n

Location: Remote

\n

Salary: Up to 240k base + 20% bonus

\n

\n

Responsibilities:

\n

\n
• Ship end-to-end ML solutions: research to modeling to production
\n
• Build statistical frameworks (balancing, normalization, paneling, cohorting)
\n
• Develop predictive models for spend forecasting, propensity, churn, and behavioral embeddings
\n
• Create causal measurement systems (synthetic controls, uplift, incrementality)
\n
• Apply privacy preserving ML (DP, cleanrooms, aggregation thresholds)
\n
• Drive production pipelines and governed model outputs
\n
• Mentor senior/staff scientists and represent methodology to stakeholders
\n

\n

\n

Requirements:

\n

\n
• 7+ years in ML/statistics with deep hands-on leadership (financial services preferred)
\n
• Strong foundation in weighting, calibration, bias correction
\n
• Expertise in supervised learning, forecasting, behavioral/tabular data
\n
• Causal inference experience (synthetic control, uplift, incrementality)
\n
• Production engineering skills (Python, SQL, cloud data warehouses)
\n
• Experience with large scale ML pipelines and evaluation frameworks
\n
• Excellent communication and methodological rigor
\n
• Representation learning / embeddings
\n
• Privacy preserving ML and cleanroom workflows
\n

\n

\n

Preferred:

\n

\n

\n
• Credit card or financial transaction data experience
\n

\n

\n

Keywords: ML Scientist, Statistical Modeling, Predictive Modeling, Paneling, Normalization, Cohorting, Causal Inference, Synthetic Control, Uplift Modeling, Transaction Data, Behavioral Modeling, Forecasting, Representation Learning, Privacy Preserving ML, Cleanrooms, Production ML Pipelines