Towards Data Science

2 min read Original article ↗

  • A step-by-step 1D CNN for text, built in Excel, where every filter, weight, and decision…

  • Understanding the process behind agentic planning and task management in LangChain

Latest

  • A data scientist’s guide to population stability index (PSI)

  • A brief overview of the math behind the Harsanyi Dividend and a real-world application in…

  • Understanding text embeddings through simple models and Excel

  • Gradient descent in function space with decision trees

  • From Random Ensembles to Optimization: Gradient Boosting Explained

  • For the last couple of years, a lot of the conversation around AI has revolved…

  • A story about failing forward, spheres you can’t visualize, and why sometimes the math knows…

  • How to make LLMs reason with verifiable, step-by-step logic (Part 1)

  • An LLM that’s 41× more efficient and 9× faster than today’s standard models

Editor’s Picks

  • Comparing metrics across datasets and models

  • Photo by Vardan Papikyan on Unsplash

    What do regular technology users think (and know) about AI?

  • From a weekend chore to a fun application of valuable operations research principles

  • Here’s how to detect point anomalies within each series, and identify anomalous signals across the…

  • When indexing hurts more than it helps: how we realized our RAG use case needed…

  • Deep work, over-identification, sports, and blogging

  • Beyond writing code, these are the design-level decisions, trade-offs, and habits that quietly separate senior…

  • Diagram of a transformer attention block showing possible gating positions (G1–G5). The image highlights G1—an element-wise sigmoid gate after SDPA—as the most effective placement. Key research questions from the Qwen gating paper are listed on the right.

    This one little trick can bring about enhanced training stability, the use of larger learning…

  • A step-by-step tutorial that explores the Python Turtle Module

The Variable Newsletter

  • Don’t miss our most popular articles of the past year!

  • Gain new insights into the art of optimizing your data science and ML workflows

Deep Dives

  • Image via Pixabay, Public Domain

    Using Agentic AI prompts with the Artificial Bee Colony algorithm to enhance unsupervised clustering and…

  • Best practices for data quality, retrieval design, and evaluation in production RAG systems

  • With concrete examples of using AI Studio Build mode to learn faster, prototype smarter, communicate…

  • An optimal solution to the well-known NP-complete problem, when the input values are close enough…

  • LLM-as-a-Judge, regression testing, and end-to-end traceability of multi-agent LLM systems

  • Instead of starting with margins and geometry, this article builds the Support Vector Machine step…