Every course I teach aims at a moment I remember having as a student: the evening you realize you can collect and analyze the data yourself. I teach statistics, machine learning, and data science — in Python and R — as working instruments, in courses built around real datasets and reproducible workflows. Students learn when to reach for a method, how to implement it, and how to explain the result to someone who will never read the code.

Ethics is course content, not an afterthought: I teach the COMPAS recidivism case so students see for themselves how a model can be statistically accurate and unjust at once. And my research keeps the classroom honest — the examples come from live projects, so students work on problems where the answer is not in the back of the book. The substantive work is under Research.

What I Teach

The arc runs from introductory statistics to modern machine learning and language models:

  • Introductory & applied statistics — descriptive statistics, probability and distributions, sampling and estimation, hypothesis testing, correlation and regression
  • Regression & causal inference — OLS, the linear probability model, causal inference from randomized experiments and observational data
  • Machine learning — supervised learning, model evaluation, and fairness and bias mitigation
  • Data science & computational methods — data wrangling, reproducible workflows, network analysis, text analysis and NLP
  • Large language models — transformers and embeddings, LLM applications, prompt engineering, retrieval-augmented generation

Experience