Towards Data Science

How should you handle outliers and missing valaues in credit-scoring models? @jumbong_junior continues his ongoing series with a new, hands-on deep dive. towardsdatascience.com/building-robus…

What factors should you prioritize when measuring the value of AI projects? Janna Lipenkova presents a detailed framework that doesn't neglect efficiency, but also looks beyond it to other crucial considerations. towardsdatascience.com/how-to-measure…

Follow along Mostafa Ibrahim's practical guide to agentic RAG pain points, and learn why these systems often fail silently in production — and how to detect these issues in time. towardsdatascience.com/agentic-rag-fa…

After a very hit-or-miss start a couple of years ago, AI coding tools are clearly here, and leaving their mark on entire professions and workflows. Stephanie Kirmer unpacks the implications of this dramatic, rapid shift. towardsdatascience.com/the-new-experi…

Curious to learn about two-stage hurdle models and their benefits for business predictions, among other use cases? Derek Tran explores their inner workings and shows how to implement them in a real-world scenario. towardsdatascience.com/two-stage-hurd…

Should your RAG system retrieve once or adapt as it goes? In this practical guide, Mostafa Ibrahim explains the shift from linear pipelines to agent-driven retrieval loops. towardsdatascience.com/agentic-rag-vs…

AI can write a lot of code in seconds. The challenge is making sure that code holds up in real applications. @EivindKjos walks through practical steps to make coding-agent output production ready. towardsdatascience.com/how-to-create-…

"So let’s walk through how this space has evolved, how to plan before using AI, why judgement and taste still matter, which AI coding tools are winning, and where the bottlenecks still are." Ida Silfverskiöld presents a detailed, experience-based breakdown of the current state of vibe coding. towardsdatascience.com/the-basics-of-…

For a practical guide to caching layers across the RAG pipeline, from query embeddings to full query-response reuse, follow along @m_mouschoutzi's latest article. towardsdatascience.com/beyond-prompt-…

If you're relying on AI coding assistants to accelerate the development process, you should still stay in control and build reliable, production-ready software. Partha Sarkar shares a detailed breakdown of best practices to follow. towardsdatascience.com/vibe-coding-wi…

For an accessible, visual guide to vectors and projections, don't miss @Nikhil_on_ds's beginner-friendly resource. towardsdatascience.com/linear-regress…

Production RAG isn’t just about retrieval quality. It’s about efficiency at scale. Partha Sarkar breaks down caching strategies that help minimize repeated LLM calls and reduce operational costs. towardsdatascience.com/zero-waste-age…

Rapid code generation is powerful, but production systems demand more than just working prototypes. @EivindKjos shares how to use Claude Code to build stronger, more dependable software. towardsdatascience.com/how-to-create-…

"Most neuro-symbolic systems inject rules written by humans. But what if a neural network could discover those rules itself?" Emmimal P Alexander presents her latest work in a thorough yet accessible deep dive. towardsdatascience.com/how-a-neural-n…

Curious about the recently announced Gemini Embeddings 2 Preview model? @taupirho fills us in on its strengths, limitations, and performance across a range of multimodal use cases. towardsdatascience.com/introducing-ge…

@r4plh @slowdownisha @Medium Happy to report we do, in fact, support LaTeX on our independent site! #shortcode" target="_blank" rel="nofollow noopener">towardsdatascience.com/how-to-format-…

@slowdownisha @Medium @TDataScience and I ask why it is not there..

Imagine being the most popular blogging platform and still no latex support!🤡

There are many advantages to self-hosting LLMs — from privacy and cost benefits to additional customization options. There are also risks. Sam Black's deep dive is a powerful resource to help you make informed decisions. towardsdatascience.com/self-hosting-y…

In his new guide to using AI coding agents effectively, @EivindKjos zooms in on a crucial step that many practitioners ignore: reviewing the outputs. towardsdatascience.com/how-to-effecti…

Don't miss the latest from Derek Tran: a new deep dive that explains the theory behind hurdle models, provides a working Python implementation, and discusses the practical considerations that matter when deploying these models in production. towardsdatascience.com/two-stage-hurd…

"I want to think about the implications of this technology on how we do our jobs both now and in the future." Stephanie Kirmer presents a thoughtful and nuanced analysis on the current state of AI coding assistants. towardsdatascience.com/the-new-experi…