Work That Speaks for Itself

Colgate-Palmolive

Bayesian Cannibalization Analysis for New Product Launches

Quantified how much of each new Colgate product's sales came from competitors versus existing Colgate products, giving the portfolio team a clear picture of true growth across five years of launches.

Colgate-Palmolive

Bayesian Shelf Assortment Optimization

Built Colgate an optimization system that recommends which products to stock at each retailer and in what mix, outperforming their existing commercial tool.

Colgate-Palmolive

Synthetic Consumer Research via Semantic Similarity Rating

Built and validated a method for running consumer surveys with AI participants that matched real human responses 90% of the time — turning weeks of research into hours.

Takeda

Bayesian State Space Modeling for CAR-NK Cell Therapy Manufacturing

Modeled a 28-day cell therapy manufacturing pipeline so Takeda could analyze completed batches, predict quality mid-process, and optimize production settings — all from just 17 donor samples.

Roche

Scalable Hierarchical Bayesian Models for Pharmaceutical-Scale Genomic Data

Proved that Bayesian analysis scales to pharmaceutical datasets by fitting a 34,000-parameter model on 250,000 observations in just over an hour using GPU acceleration.

Supercell

Bayesian MMM for Mobile Game User Acquisition

Helped Supercell understand which marketing channels actually drive mobile game player acquisition — and where to shift budget for better returns.

Real Madrid Football Club

Fan CLV Modeling with PyMC-Marketing

Built models to estimate the long-term value of individual fans, helping one of the world's largest sports franchises make smarter retention and engagement decisions.

Haleon

Bayesian Modeling Advisory and Team Upskilling for Consumer Healthcare

Coached Haleon's internal data science team through the hard parts of building their own Bayesian marketing models — from missing data to GPU acceleration.

Indigo Ag

Spatial Gaussian Process Modeling for Agricultural Treatment Effects

Separated real microbial treatment effects from the natural variation in agricultural fields, giving Indigo Ag a defensible measure of whether their seed treatments actually improve crop yields.

Syngenta

Expert Review of Bayesian Dose-Response Models for Agrochemical Research

Helped Syngenta's data science team sharpen their probabilistic models for measuring chemical potency in crop protection research.

Akili Interactive

Bayesian Cognitive Assessment Scoring for Digital Therapeutics

Developed rigorous scoring methods for cognitive assessments and accelerated inference on complex models of cognition, supporting clinical research behind an FDA-approved digital therapeutic for ADHD.

Erisyon

Bayesian HMM for Single-Molecule Protein Sequencing

Built a probabilistic pipeline that decodes noisy fluorescence signals to identify individual proteins, delivering calibrated confidence estimates at the throughput Erisyon's technology demands.

Everysk

Bayesian Private Equity Index from Capital Cash Flows

Uncovered private equity and venture capital performance from cash flow data alone, producing a return index that tracks established industry benchmarks.

VisualVest

Probabilistic Customer Lifetime Value for a Robo-Investment Platform

Developed a probabilistic model to forecast customer lifetime value for a robo-investing platform, where off-the-shelf approaches couldn't account for variable, AUM-based revenue.

Appodeal

Bayesian MMM for Mobile App Developer Acquisition

Identified which marketing channels were actually driving developer acquisition for Appodeal's platform, with honest uncertainty around each channel's return.

Live Nation

Hierarchical Bayesian MMM for Concert Tour Ticket Sales Across 125+ Artists

Built a marketing effectiveness model for concert ticket sales that handles the unique rhythm of tour promotion — from announcement buzz through on-sale surges — across more than 125 artists.

Fox Broadcasting Company

Show-Level MMM and NFL Promo ROI Analysis

Helped Fox Broadcasting's data science team build marketing models for TV advertising revenue and measure the true impact of NFL promotional campaigns.

Gain Theory

Hierarchical Bayesian MMM and Bass Diffusion for a WPP Marketing Science Consultancy

Worked with Gain Theory across multiple fronts — strengthening their Bayesian modeling practice, delivering marketing models for their clients, and building a viewership forecasting model for streaming content.

Fabletics

MMM Upgrade from PyMC3 to PyMC-Marketing with HSGP

Guided the Fabletics data science team through a major model modernization after their lead modeler departed, upgrading their marketing mix model and adding time-varying parameters.

Swarovski

Bayesian MMM with Time-Varying Intercept for Luxury Retail

Improved prediction accuracy by 20% for Swarovski's marketing mix model by letting the baseline sales level shift naturally over time instead of holding it fixed.

L.L. Bean

Building In-House Hierarchical MMM Capability Across 50 US DMAs

Helped L.L. Bean's team build their own marketing mix model spanning 50 US markets with catalog attribution, replacing a vendor model they couldn't inspect or modify.

Wegmans Food Markets

Bayesian New Store Site Selection and Cannibalization Modeling

Built a model that jointly predicts new store sales and quantifies how much nearby existing stores would lose, with 13-14% prediction accuracy and a measured cannibalization effect around 1%.

Los Angeles Dodgers

Time Series Modeling Advisory

Provided expert guidance on time series modeling for the Los Angeles Dodgers analytics team through the Expert Access Program.

SALK

Multilevel Regression & Post-Stratification for Public Opinion Polling

Helped an Estonian civic data consultancy extract reliable public opinion insights from sparse survey data, producing stable estimates even for hard-to-reach demographic groups.

Large Video Streaming Service

Bayesian A/B Testing at Scale (100M+ Observations)

Made Bayesian A/B testing practical at massive scale — tests that previously couldn't finish now run in 22 seconds on a standard desktop, even with 100 million observations.

Alva Labs

Bayesian Psychometric Modeling for Hiring Assessment Validity

Built psychometric and validity models that substantially outperformed Alva Labs' existing assessment system, giving them both better trait estimates and hard evidence that their assessments predict job performance.