Dylan Poulsen Consulting

Thoughtful statistical consulting for clear, defensible decisions

Applied Data Science and Bayesian Causal Inference Consulting

Careful analysis for decisions that matter.

I help teams build clear, defensible analyses and models through principled workflow, Bayesian methods, and thoughtful communication.

Book a Consultation Explore Services

As featured in

  • The New York Times
  • The Pudding
  • Daily Mail

Consulting shaped by research, teaching, and practice

I bring a research and teaching background in data science and mathematics to consulting work that needs both rigor and judgment.

  • Clear and defensible analyses
  • Principled workflow from question to delivery
  • Bayesian causal inference when it serves the problem
  • Shiny apps, reports, and machine learning models

What I Help With

I work where practical constraints meet analytical rigor: causal questions, forecasting, noisy operational data, technical reporting, and workflows teams can run repeatedly.

Bayesian Causal Inference

Uncertainty-aware analysis for intervention questions, experimental design, observational studies, and decision support.

Applied Data Science Systems

Data cleaning, visualization, Shiny apps, machine learning models, and analysis pipelines for real operational or research data.

Bayesian Workflow And Reporting

Quarto reports, principled model-building workflow, and repeatable analysis systems that preserve structure and reduce repetitive labor.

Ethical Analytics Strategy

Reproducible, interpretable work with documented assumptions, honest uncertainty, and respect for the people in your data.

Applied Data Science Leadership

I lead applied data science work in both consulting and academic settings, with a focus on reproducible systems that support real decisions under uncertainty.

As Director of the Data Science Innovation Lab at Washington College, I also supervise project-based analytics work that sharpens the same delivery standards I use in industry engagements.

How Ambiguity Becomes Delivery

Bayesian workflow for client work

  • Scope the question: define a usable objective, data requirements, and success criteria.
  • Build the workflow: clean data, check assumptions, fit interpretable models, and deliver Shiny apps or reports stakeholders can actually use.
  • Ship decisions: deliver reproducible reports, clear recommendations, and next-step options.

This is the same process I use in client consulting engagements, from first conversation to final handoff.

Selected Proof

Machine Learning For RNA Quality

With Thermo Fisher, I developed a machine learning model to assess RNA quality using Thermo Fisher equipment, contributing to a conference poster presentation.

View poster

The Onion Constant

My work on the mathematically optimal way to cut an onion turns a culinary question into a public-facing optimization problem. It has been cited by The New York Times, The Pudding, Daily Mail, and outlets around the world.

Read the analysis

Teaching Modern Data Science

I design and deliver technical training in Bayesian methods, causal inference, and reproducible analytical workflow for teams that want stronger in-house capability.

Learn about my background

Have A Data Problem That Needs Clear Thinking?

I bring careful methods, practical communication, and transparent uncertainty to every consulting engagement.

Book a Consultation