Towards Data Science

2 min read Original article ↗

  • In my latest posts, we’ve talked a lot about prompt caching as well as caching…

  • Using Codex and MCP to connect Google Drive, GitHub, BigQuery, and analysis in one real workflow

Latest

  • Split illustration of a librarian and a friendly robot in a library. On the left, the robot is overwhelmed by messy stacks of books and papers labelled “irrelevant.” On the right, the librarian hands the robot a small, tidy stack of relevant books, showing a more focused and efficient research process.

    Why retrieval that looks excellent on paper can still behave like noise in real RAG…

  • My last article was about implementing Like-for-Like (L4L) for Stores. After discussing my solution with…

  • Proactivity, blocking, and planning

  • Photo by Jeanna Song on Unsplash

    Understanding how to set up human-in-the-loop (HITL) agentic workflows in LangGraph

  • Data Leakage, Real-World Models, and the Path to Production AI in Healthcare

  • Supercharge Claude Code with continual learning

  • How AI agents, data foundations, and human-centered analytics are reshaping the future of decision-making

  • We’ve become remarkably good at building sophisticated agent systems, but we haven’t developed the same…

  • How to leverage a framework to effectively prioritize AI Initiatives to rapidly accelerate growth and…

Editor’s Picks

  • Rapid prototyping with Replit, AI agents, and minimal manual coding

  • A step-by-step guide to making your OpenAI apps faster, cheaper, and more efficient

  • Building products without the coding part

  • A practical guide to caching layers across the RAG pipeline, from query embeddings to full…

  • Accelerate coding with AI while staying in control and building reliable, production-ready software.

  • Why one model can’t do two jobs

  • Photo by Lan Gao on Unsplash

    The seduction of AI code assistants

  • Privacy. Cost. Customization. Everything you need to know—step by step.

  • It’s a feature of the architecture

The Variable Newsletter

  • Authors can now benefit from updated earning tiers and a higher article cap

  • Sorting through the good, bad, and ambiguous aspects of vibe coding

Deep Dives

  • Concept drift in fraud detection model causing domino effect of fraudulent transactions with early anomaly detection signal

    This Article asks what happens next. The model has encoded its knowledge of fraud as…

  • Piecewise linear approximations are a practical way to handle nonlinear constrained models using LP/MIP solvers…

  • Handling outliers and missing values in borrower data using Python.

  • While efficiency is an important source of AI value, it is only part of the…

  • A visual guide to vectors and projections

  • A hands-on case study and practical guidance