Blog
The Cox survival guide: hazards, models and time
Abstract
This note is an overview of survival analysis with a focus on the Cox proportional hazards model, a key tool for modelling time-to-event data. We define hazard functions, address right-censoring, and derive estimators like Nelson-Aalen and Kaplan-Meier. The Cox model's semiparametric form and partial likelihood estimation are discussed, along with evaluation metrics such as concordance probability.
Intro to Jax
Abstract
Mutiple packages came to solve the speed issue of python in scientific computing. The de facto standard is of course numpy. Think also numba, jax, torch or even a new language 
that go beyond what numpy offers. This post is a quick intro to my personal preference as of Jan 2025.
Mini-tutorial for python packaging, release and publish
Abstract
This mini-tutorial is a beginner's cheatsheet to python packaging. Check Python packaing user guide for an authoritative guidance on the topic.
Essentials of polars for pandas experts
Abstract
pandas is a standard tool for every data professional, although it does not scale well in production.
Yet, being a standard is a strategic position to be, as libraries coming to solve the scale issue tend to meet data professionals where they are, by mimicing the pandas API (think: dask, pyspark.pandas).
polars is a new-ish tool that is probably replacing pandas at the time of writing. The goal of this post is to introduce the kind of mindset change needed to fully exploit polars in production.