Towards Data Science
I Replaced a 15-Minute Booking Process with a LangGraph AI Agent
Artificial Intelligence I Replaced a 15-Minute Booking Process with a LangGraph AI Agent A step-by-step guide to…
ai · source
Towards Data Science on WeSearch
Recent ai headlines from Towards Data Science.
Lead story
Towards Data Science
How to Apply Coding Agents to Non-Programming Tasks
Agentic AI How to Apply Coding Agents to Non-Programming Tasks Perform non-programming tasks with coding agents Eivind Kjosbakken Aug 2, 2026 12 min read Share In this article, I’ll discuss how to perform…
The latest
Towards Data Science
Coding Agents Don’t Need Bigger Context Windows — They Need a Context Compiler
Coding agents often waste a large portion of their context windows on irrelevant code, reducing their efficiency. A…
Towards Data Science
Put the Agent Inside the Workflow
LLM Applications Put the Agent Inside the Workflow A hybrid LLM application pattern that combines a predefined…
Towards Data Science
The 3× Token Bill We Didn’t See Coming
Agentic AI The 3× Token Bill We Didn’t See Coming How a seemingly harmless move to a multi-agent architecture quietly…
Towards Data Science
When the Code Becomes the CEO: Why Your Next Manager Might Be a Decentralized Agentic Loop
Artificial Intelligence When the Code Becomes the CEO: Why Your Next Manager Might Be a Decentralized Agentic Loop In…
Towards Data Science
How to Debug AI Coding Agents When They Change the Wrong Thing
Agentic AI How to Debug AI Coding Agents When They Change the Wrong Thing A practical tutorial for recording model…
Towards Data Science
How Benders Decomposition Works Part I: Optimality Cuts
Operations Research How Benders Decomposition Works Part I: Optimality Cuts A friendly introduction to one of the most…
Towards Data Science
The Python Ecosystem That Changed AI Development
Machine Learning The Python Ecosystem That Changed AI Development How one open-source ecosystem made state-of-the-art…
Towards Data Science
How to Organize All of Your Coding Agent Tasks
Agentic AI How to Organize All of Your Coding Agent Tasks Optimise how you interact with your coding agents Eivind…
Towards Data Science
How to Build a Context Layer and a Company Brain
Machine Learning How to Build a Context Layer and a Company Brain What it actually takes to turn a company's scattered…
Towards Data Science
A Simplified View of the Jacobian Conjecture
Math A Simplified View of the Jacobian Conjecture The full conjecture is stated over abstract fields, but the…
Towards Data Science
How to Decode the Temperature Parameter in LLMs
Large Language Models How to Decode the Temperature Parameter in LLMs How statistical physics explains the transition…
Towards Data Science
Prompt Engineering Is Solved—Prompt Management Isn’t
It’s what inevitably happens when a prompt’s input variables change and zero tooling checks the actual call sites…
Towards Data Science
Why Your Best Predictive Model Gives the Wrong Treatment Effect
Data Science Why Your Best Predictive Model Gives the Wrong Treatment Effect Why prediction-driven variable selection…
Towards Data Science
Los Movimientos, Part II: Solving Large Pickup-and-Delivery Problems with Adaptive Large Neighborhood Search
The article discusses using Adaptive Large Neighborhood Search (ALNS) to solve large-scale pickup-and-delivery routing…
Towards Data Science
Avoiding Entity Key Drift in a Data Lake: Step 1, Normalization
Assigning a primary key to an identifier string the first time you observe it is a reasonable starting point. An…
Towards Data Science
How Much Does a Local LLM Actually Cost to Run? I Measured Every Watt on Apple Silicon
Large Language Models How Much Does a Local LLM Actually Cost to Run? I Measured Every Watt on Apple Silicon Five…
Towards Data Science
MCP Explained: How Modern AI Agents Connect to the Real World
We saw how the model generates a tool call, our code executes it, and the result gets fed back to the model. This…
Towards Data Science
Don’t Just “Throw Adam at It”: Misunderstanding Adam Will Cost You
Machine Learning Don’t Just “Throw Adam at It”: Misunderstanding Adam Will Cost You You "vibe-coded" the import.…
Towards Data Science
Backpropagation Explained for Beginners (Part 2): There Has to Be a Better Way
Deep Learning Backpropagation Explained for Beginners (Part 2): There Has to Be a Better Way The idea that makes…
Towards Data Science
“Los Movimientos”: The Routing Problem That Nearly Broke My Spirit
Operations Research “Los Movimientos”: The Routing Problem That Nearly Broke My Spirit Using mathematical optimization…
Towards Data Science
Reducing Human Annotation with ML Active Learning
Machine Learning Reducing Human Annotation with ML Active Learning In a world where human time is expensive, learn how…
Towards Data Science
The Most Beautiful Statistic: The History and the Science of the Humble Mean
The first month was spent in orbit around the martian planet and on July 20, 1976 Viking Lander 1 separated from the…
Towards Data Science
How I Reproduced BM25, Dense Retrieval, and SPLADE on a 16GB MacBook
Large Language Models How I Reproduced BM25, Dense Retrieval, and SPLADE on a 16GB MacBook A practical reproduction of…
Towards Data Science
How to Efficiently Prompt Claude Code
LLM Applications How to Efficiently Prompt Claude Code Maximize your efficiency with Claude Code Eivind Kjosbakken Jul…
Towards Data Science
How to Give an LLM Agent a Browser
LLM Applications How to Give an LLM Agent a Browser Building a browser-use agent with OpenAI Agents SDK and Playwright…
Towards Data Science
How to Optimize Vector Search When RAM Gets Too Expensive: On-Disk vs. In-Memory ANN Indexes
Machine Learning How to Optimize Vector Search When RAM Gets Too Expensive: On-Disk vs. With the rise of agentic…
Towards Data Science
The Fluid Simulator That Doesn’t Solve the Fluid Equations
You’ve seen this pattern trailing behind bridge pillars, airplane wings, and chimney stacks. It’s one of the most…
Towards Data Science
Lessons Learned After 8.5 Years of ML
Machine Learning Lessons Learned After 8.5 Years of ML Patience, Optimism, Discipline, Projects, Teams Pascal Janetzky…
Towards Data Science
When Data Science Makes Us Sad: The Story of an Overbooked Flight
Data Science When Data Science Makes Us Sad: The Story of an Overbooked Flight $8 million vs $5k + Potentially Going…
Towards Data Science
Tabular LLMs: An Introduction to the Foundation Models That Predict Your Spreadsheet
Large Language Models Tabular LLMs: An Introduction to the Foundation Models That Predict Your Spreadsheet A…
Towards Data Science
Build and Run an Intelligent Document Processing (IDP) System in the Cloud
The system was designed to allow Freedom of Information-type requests from Irish citizens wanting to know whether…
Towards Data Science
Loop Engineering for RAG Generation: An LLM Cascade from a Cheap Local Model Up to a Hosted Flagship
You have a few hundred typed fields to pull out of a stack of real documents: amounts, dates, coverage limits, one…
Towards Data Science
Context Windows Forget What Matters — I Built a Usage-Reinforced Decay Engine for AI Agent Memory
Emmimal P Alexander Jul 24, 2026 23 min read Share Image by the author, generated with ChatGPT (DALL·E) TL;DR: Most…
Towards Data Science
Most RAG Hallucinations Are Extraction Errors: Seven Patterns for a Typed Generation Contract
Large Language Model Most RAG Hallucinations Are Extraction Errors: Seven Patterns for a Typed Generation Contract…
Towards Data Science
Prompt Engineering Isn’t Enough: How Four Bricks of Context Engineering Stop RAG Hallucinations
Large Language Models Prompt Engineering Isn’t Enough: How Four Bricks of Context Engineering Stop RAG Hallucinations…
Towards Data Science
Loop Engineering with Adaptive Parsing in Action: Parsing Flat Tables with Azure and Figures with a Vision LLM
The article describes an adaptive parsing strategy for enterprise document intelligence that starts with a cheap…
Towards Data Science
Water Cooler Small Talk, Ep. 12: Byzantine Fault Tolerance
The concept of Byzantine Fault Tolerance allows a system to operate properly even when malicious actors are present.…
Towards Data Science
Automatically Assign a Category to Uncategorized Rows in Power Query and DAX
Data Engineering Automatically Assign a Category to Uncategorized Rows in Power Query and DAX Sometimes we must assign…
Towards Data Science
How to Get the Most Out of Claude Fable 5
LLM Applications How to Get the Most Out of Claude Fable 5 Maximize your Claude Fable 5 usage Eivind Kjosbakken Jul…
Towards Data Science
Stop Returning Text from RAG: The Typed Answer Contract That Prevents Hallucination
The article discusses the concept of a typed answer contract in large language models to prevent hallucination. This…
Towards Data Science
Setting Up Your Own Large Language Model
The article discusses setting up a large language model on a personal laptop, highlighting the benefits of having a…
Towards Data Science
Long Context vs. Short Context Model: When Does a Long Context Model Win?
The article discusses the trade-off between using long context models and short context models in artificial…
Towards Data Science
AI Agents Explained: What Is a ReAct Loop and How Does It Work?
The article discusses the concept of a ReAct loop in AI agents, which allows them to reason, act, and observe their…
Towards Data Science
LLM Wikis Are Over-Engineered — I Replaced Mine With a Pure Python Compiler
The author replaced their Large Language Model wiki with a pure Python compiler to structure local markdown notes.…
Towards Data Science
4 Lines You Should Include in Your Claude Skill
Agentic AI 4 Lines You Should Include in Your Claude Skill Without these, Claude might end up being confidently wrong.…
Towards Data Science
How to Refactor Code with Claude Code
LLM Applications How to Refactor Code with Claude Code Improve coding agent productiveness with refactored code Eivind…
Towards Data Science
I Spent May Evaluating Different Engines for OCR
The article discusses an evaluation of various OCR engines conducted by the author, testing fourteen different engines…
Towards Data Science
Why AI Is NOT Stealing Your Job
The article discusses the impact of artificial intelligence (AI) on the job market, emphasizing that AI itself is not…
How WeSearch handles this source
WeSearch's declared handling of Towards Data Science's content. Indexing, snippets, summaries, retrieval and training are separate questions — see the rights registry or read this source's machine-readable record.
Indexing: Allowed
Snippet: Allowed
AI summary: Limited
Retrieval / RAG: Not asserted
Model training: Not asserted
Commercial reuse: Not permitted