Note: The job is a remote job and is open to candidates in USA. Blackstraw.ai is an end-to-end technology services company specializing in Artificial Intelligence and Engineering solutions. They are seeking a Lead Data Scientist to build and productionize marketing mix and pricing models that quantify the sales impact of various factors across products and stores, while also managing a team of data scientists.

Responsibilities

• Design and estimate sales response models incorporating price, TPR, merchandising, promotional mechanics, seasonality, competitor cross-effects, and price/promotion thresholds, at store x product x week granularity
• Apply Bayesian hierarchical and panel modeling techniques to pool information across store panels, balancing degrees-of-freedom constraints against store-level heterogeneity; implement mean-scaling transformations to stabilize elasticity estimates
• Build and calibrate media transformation pipelines (adstock decay, Adbudg saturation curves) to estimate incremental sales, inflection points, and saturation levels for media investment
• Build distributed, production-grade estimation pipelines in Python/PySpark that scale across large panels of stores and products with 2+ years of weekly history
• Present model outputs, elasticities, and simulation results to pricing, trade, and media stakeholders; make investment and pricing recommendations grounded in the models
• Manage and mentor a team of data scientists; define modeling standards, QA/validation protocols, and the technical roadmap for the MMM platform

Skills

• Master's degree in Statistics, Econometrics, Applied Mathematics, Computer Science, or a related quantitative field (PhD preferred)
• Minimum 4 years of experience building statistical or econometric models (regression, time-series, or panel data) as a Data Scientist or Statistician
• Minimum 8 years of experience in machine learning model development and MLOps, including feature engineering, model training/validation pipelines, CI/CD for models, containerization, versioning, and monitoring in production
• Demonstrated expertise in Bayesian statistics (hierarchical/multilevel models, MCMC, prior specification) and classical econometrics (panel/fixed-effects, GLS)
• Proficiency in Python (statsmodels, PyMC/Stan, scikit-learn) and PySpark for large-scale, distributed model estimation
• Experience building Marketing Mix Models (MMM): price/promotion elasticity, adstock and diminishing-returns (Adbudg) media response curves, store panel clustering, and mean-scaling transformations
• Experience with SQL and cloud data platforms (BigQuery, Snowflake, Databricks) on multi-TB retail/POS datasets
• Ability to translate model coefficients — elasticities, response curves, ROI — into business recommendations for pricing, trade, and media investment

Company Overview

Company H1B Sponsorship