All Posts

Introduction into ONNX, an open-source specification that allows us to persist the state of a trained model as well as its inference logic.

Estimating Conditional Average Treatment Effects with the metalearners library

We show how we ship conda environments into production using the pixi package manager. To do that, we introduce a tool called pixi-pack.

We built a tool to automatically keep all of our repositories up-to-date with the latest best practices and dependency versions.

The new glum release introduces a formula interface that simplifies preprocessing and offers various enhancements for categorical and mixed sparse and dense data.