Never miss a new edition of variablea weekly newsletter featuring a top-notch selection of editor’s picks, details, community news, and more.
Like many LLM-based workflows before it, Vibe Coding attracted strong opposition and sharp criticism not because it lacked value, but because of unrealistic and hype-based expectations.
The idea of leveraging powerful AI tools to experiment with app building, generate quick prototypes, and iterate quickly is hard to argue with. The problem typically begins when a human practitioner receives the output produced by a model and assumes it to be robust and error-free.
To help you sort through the good, bad, and ambiguous aspects of vibe coding, we consulted experts. The lineup we have for you this week provides a nuanced and practical look at how AI code assistants work, when and how to use them.
Unbearable lightness of coding
“The amount of technical questions weighs more heavily on my shoulders than ever before.” In her powerful and brutally honest Confessions of a Vibe Coder, Elena Jolkver takes an unflinching look at what it meant to be a developer in the era of Cursor, Claude Code, and more. She also argues that the way forward requires recognizing both the speed and productivity benefits of vibecoding. and Its (many) potential pitfalls.
How to run cloud code for free using Ollama’s local and cloud models
If you’re already sold on the promise of AI-assisted coding, but are worried about its steep costs, don’t miss Thomas Reid’s new tutorial.
How cursors actually index into the codebase
Curious about the inner workings of the most popular vibe coding tool? Kenneth Leung details the Cursor RAG pipeline, which ensures efficient indexing and retrieval for coding agents.
Most read articles this week
In case you missed it, here are three articles that resonated with a wide range of readers over the past week.
Beyond the Context Window: Recursive Language Models in Action, by Mariya Mansurova
We explore practical approaches to analyzing large datasets using LLM.
Causal ML for Aspiring Data Scientists, by Ross Lauterbach
An easy-to-understand introduction to causal inference and ML.
Vector Search Optimization: Why You Need to Flatten Structured Data, by Oleg Tereshin
An analysis of how flattening structured data can improve precision and recall by up to 20%.
Other recommended books
Python skills, MLOps, and LLM assessments are just some of the topics covered in this week’s top-notch articles.
Why SaaS Product Management will be the domain of choice for data-driven professionals in 2026 by Yassin Zehar
Creating an Etch A Sketch App with Python and Turtle by Mahnoor Javed
Machine Learning in Production? What This Really Means, by Sabrin Bendimerad
Evaluating Multi-Step LLM-Generated Content: Why Your Customer Journey Needs Structural Metrics, by Diana Schneider
Google Trends is misleading: How to use Google Trends data for machine learning by Leigh Collier
Introducing new authors
Please take the time to see the great work of TDS contributors who have recently joined the community.
- Luke Stucky looked at how neural networks approach the problem of music similarity in the context of recommendation apps.
- Aneesh Patil described a geospatial data project aimed at estimating neighborhood-level pedestrian risk.
- Tom Narock argues that the best way to address data science’s “identity crisis” is to reframe data science as an engineering practice.
We love publishing articles by new authors, so if you’ve recently written an interesting project walkthrough, tutorial, or theoretical reflection on one of our core topics, why not share it with us?
