Shadowgraph

Tech stack & purpose

CausalML is a Python package developed by Uber that provides uplift modeling and causal inference methods using machine learning algorithms. The package enables users to estimate the Conditional Average Treatment Effect (CATE) from experimental or observational data, with typical applications in campaign targeting optimization and personalized engagement strategies. The project is built as a standard Python library with support for various machine learning backends, including a JAX/Flax implementation of the Causal Effect Variational Autoencoder (CEVAE) alongside existing PyTorch implementations. The codebase draws on recent academic research in causal inference and treatment effect estimation, with the project currently stable but incubated for long-term support.