Never miss a new edition of variablea weekly newsletter featuring a top-notch selection of editor's picks, details, community news, and more.
Yes, it's 2026. TDS is already focused on an eventful year of growth and learning. We published a lot of great articles last month, including right in the middle of the holiday season. Don't miss any of them.
This week, we're dedicating Variables to one last hurray for 2025 and highlighting some of December's most popular articles. But don't get me wrong. It covers timely and practical topics in machine learning, data science, and AI that will remain relevant in the weeks and months ahead.
GraphRAG in action: How to build cost-effective, high-recall search systems
When a “standard” RAG system no longer works for you, it's time to try out the power of GraphRAG. Partha Sarkar's detailed guide is a great starting point for anyone interested in leveraging hybrid pipelines and tinkering with this powerful approach to cost savings.
6 lessons learned building a RAG system in production
For more practical insights on RAG, we highly recommend Sabrine Bendimerad's roundup of best practices covering data quality, evaluation, and more.
How to use simple data contracts in Python for data scientists
Eirik Berge provides a quick and concise guide to using the open source library Pandera to define schemas as class objects.
Other December highlights
From learning algorithms with Excel to improving Pandas performance, here are some of the most read and shared articles from last month.
Machine Learning and Deep Learning Advent Calendar Series: Blueprint, by Angela Shi
How Agent Handoff Works in Multi-Agent Systems, by Kenneth Leung
Reading Research Papers in the Age of LLM by Parul Pandey
7 Panda Performance Tricks Every Data Scientist Should Know by Benjamin Nweke
What Happens When You Build an LLM Using Only 1s and 0s (Moulik Gupta)
Introducing new authors
Please take the time to see the great work of TDS contributors who have recently joined the community.
- Jasper Schroeder shared valuable takeaways from a recently completed Advent of Code programming challenge.
- Morris Stallmann (along with co-author Sebastian Humberg) has provided a comprehensive and practical introduction to data drift (and how to detect it in a timely manner).
- Alon Lanyado focused on another challenge that data scientists and ML practitioners often face: covariance shift.
Do your New Year's resolutions include publishing with TDS and joining the Author Payments Program? It's time to submit your latest draft.
