Aleksander Molak is an independent machine learning researcher, author, consultant, educator, and entrepreneur. His work concentrates on the areas where machine learning creates the most value and the most risk: causality, natural language processing, and AI strategy. His stated mission is to translate complex concepts into understandable, bite-size pieces, a philosophy that runs through everything he does, from research to teaching and consulting.

Causality is where Aleksander has made his deepest mark. He teaches and applies techniques that help teams move beyond correlation, including counterfactual reasoning, treatment effect estimation, conditional average treatment effect (CATE) estimation, and uplift modeling with meta-learners such as the T-learner and S-learner. His toolkit also covers debiasing methods like double and debiased machine learning, refutation testing for model validation, causal discovery algorithms, and the practical question of how large language models can strengthen causal workflows, from extracting features from text to inferring unobserved confounders.

As a consultant and educator, Aleksander works with practitioners and organizations that want to make better decisions with data. He shares his ideas through writing, courses, and talks, always with the same goal: helping people ask better questions of their models and build machine learning systems that hold up in production.