Towards Data Science
Articles about Towards Data Science
We Specified an AI Agent in 2016. We Called It Analytics
Re-reading my own IBM patent on real-time concert feedback as what it actually was: an agentic loop with multimodal perception, memory-grounded inference, and a human in the loop. What it got right, and what a decade of hindsight teaches agent builders today.
I Built a Trap for AI Image Judges. They Walked Into a Different One
Three vision AI models ignored the pretty image, yet GPT-4o invented answers and Claude kept answering B.
A Practical Guide to OpenAI’s New Decisions API
Is this the JEV killer?
Agentic Systems: A Practitioner's Guide to 6 Advanced Architectural Patterns
Beyond basic ReAct loops: six production-oriented architectures for building autonomous, reliable, and scalable LLM agents. The post Agentic Systems: A Practitioner's Guide to 6 Advanced Architectural Patterns appeared first on Towards Data Science.
How Can AI Agents Read Untrusted Sources Safely?
Add architectural guardrails around how agents read untrusted sources and use them in the workflow. The post How Can AI Agents Read Untrusted Sources Safely? appeared first on Towards Data Science.
Why Temperature 0 Isn’t Deterministic
A one-line formula for how often an LLM’s top token flips, and why identical completions fall apart after about a hundred tokens
Can TypeSafe's Jev Make AI Agents Safer Without Another LLM?
How a decision-first model can catch risky tool calls before they turn into real-world actions The post Can TypeSafe's Jev Make AI Agents Safer Without Another LLM? appeared first on Towards Data Science.
Where Does the Money Go Across Long-Running Coding Agents?
The cost of control The post Where Does the Money Go Across Long-Running Coding Agents? appeared first on Towards Data Science.
Build Your First AI Agent with One Tool Call
AI Agents for Beginners, Part 1 The post Build Your First AI Agent with One Tool Call appeared first on Towards Data Science.
Stop Using AI. Start Hiring It
AI agents can now write more code than any of us can read. That changed how I think about AI. The post Stop Using AI. Start Hiring It. appeared first on Towards Data Science.
Everyone Is Selling AI at You — Here’s How to Keep Your Judgement
Three moves to decode the hype
AI Agents Beat PyTorch: Writing Faster CUDA Kernels
AI agents can now write CUDA kernels that outperform PyTorch—but proving those speedups are real is the harder problem. I put them to the test on an NVIDIA DGX Spark and found that benchmark design matters just as much as the code.
Your Model’s MSE Is Lying to You III: Time Series Diffusion
A diffusion head for probabilistic time series forecasting: why a forecast can get the mean and the variance right and still describe a future that never happens, and how to fix it.
AI Engineers, Applied Scientists, Forward Deployed Engineers: Let’s Unpack
An exploration of the new data science roles emerging after the Large Language Model explosion.
How Wrong Is Your Marketing Mix Model (MMM)?
A well-built MMM can still be wrong, because the spend it learns from never held the information it needs. This article tests whether changing when you spend, not how much, can fix that. The post How Wrong Is Your Marketing Mix Model (MMM)? appeared first on Towards Data Science.
Introduction to Reinforcement Learning: Multi-Armed Bandit Simulation in Python
When machines learn by themselves The post Introduction to Reinforcement Learning: Multi-Armed Bandit Simulation in Python appeared first on Towards Data Science.
The Consistency Quadrant: A Visual Guide to LLM Reliability
Mapping structural variance against execution outputs to estimate coding agent reliability without ground truth. The post The Consistency Quadrant: A Visual Guide to LLM Reliability appeared first on Towards Data Science.
When Do PINNs Beat Classical Numerical Methods? A 1D vs 5D Experiment
I put a physics-informed neural network up against a plain finite-difference solver twice: once in 1D and once in 5D. The winner changed.
How I Use AI to Learn New Topics Faster: An AI-Assisted Learning Framework
A practical framework for turning AI into a thinking partner The post How I Use AI to Learn New Topics Faster: An AI-Assisted Learning Framework appeared first on Towards Data Science.
A Google Team Measured Half of My Argument, and Left the Other Half Open
Towards Spec-Driven Test Automation: Part 3 The post A Google Team Measured Half of My Argument, and Left the Other Half Open appeared first on Towards Data Science.
How to Make a Cloud Read a Drone’s Mind (and Cut Data Usage by 94%)
Two copies of one small AI model keep a remote machine in sync with the cloud, with no constant streaming. A beginner-friendly world model tutorial in PyTorch.
I Hid Four Traps in a Forecasting Task. Here Is What Four AI Assistants Did
A controlled test of Gemini, DeepSeek, ChatGPT, and Claude on leakage, reporting delays, promotion effects, and structural breaks. The post I Hid Four Traps in a Forecasting Task. Here Is What Four AI Assistants Did. appeared first on Towards Data Science.
Computer Vision: SIFT algorithm (Scale Invariant Feature Transform)
Elegant object matching from several viewpoints The post Computer Vision: SIFT algorithm (Scale Invariant Feature Transform) appeared first on Towards Data Science.
How to Build a Cheap, Yet Reliable Model Router With Jev
Model routing can finally be a design choice rather than a system enhancement The post How to Build a Cheap, Yet Reliable Model Router With Jev appeared first on Towards Data Science.
How to Govern AI Agents
From guarding one agent to steering a fleet
The Reversal Curse: Why a Language Model That Knows “A Is B” Can’t Tell You “B Is A”
A model can recall a fact in the direction it learned it and fail in the other. Here's a toy version of why The post The Reversal Curse: Why a Language Model That Knows “A Is B” Can’t Tell You “B Is A” appeared first on Towards Data Science.
Measuring the Creativity Potential of LLM Agents
Trying to answer the question of "Can LLM agents discover?" through the lens of creativity The post Measuring the Creativity Potential of LLM Agents appeared first on Towards Data Science.
How to Use a PINN for a Navier-Stokes Inverse Problem
A from-scratch PyTorch build that recovers blood flow, viscosity, and wall shear stress in a narrowed artery from 40 noisy velocity readings
Where the Agent Development Lifecycle Fits
Coordinating agent capability development with the application it powers The post Where the Agent Development Lifecycle Fits appeared first on Towards Data Science.
How to Build a Control Plane for AI Agents
Giving an LLM permission to act, in 9 steps
Autoencoders vs. PCA: I Rigged the Test and PCA Still Won
A theoretical advantage that didn't survive contact with a real benchmark. The post Autoencoders vs. PCA: I Rigged the Test and PCA Still Won appeared first on Towards Data Science.
Can an Apartment Search Agent Call the Model Fewer Times and Still Find Good Matches?
I traced 2,500 listing checks with Weights & Biases Weave, removed avoidable model work one change at a time, and scored every version against the same answers. The post Can an Apartment Search Agent Call the Model Fewer Times and Still Find Good Matches? appeared first on Towards Data Science.
What the ReLU Revolution Revealed About Biological Plausibility
The activation function was never a fixed biological commitment but a working hypothesis revised under empirical pressure. The post What the ReLU Revolution Revealed About Biological Plausibility appeared first on Towards Data Science.
Your AI Bill Is a Toll Booth. Stop Paying Twice
The budget nobody saw coming
How Many Stories Can Your Data Tell?
How the way we represent data can change what we think the data is saying The post How Many Stories Can Your Data Tell? appeared first on Towards Data Science.
Insight Is Still the Currency of Data Science
Coding agents give us more time for discovery and our review practices should follow the analysis The post Insight Is Still the Currency of Data Science appeared first on Towards Data Science.
How to Solve Issues When You Nest Measures While Overwriting the Same Filter
In DAX, we reuse existing measures all the time when writing new measures. What happens when we write a measure based on another measure and try to change a filter already set in the nested measure? The post How to Solve Issues When You Nest Measures While Overwriting the Same Filter appeared first on Towards Data Science.
Towards Spec-Driven Test Automation: Part 2
One run, and what it actually proves The post Towards Spec-Driven Test Automation: Part 2 appeared first on Towards Data Science.
I Compacted 1,000 Apache Iceberg Files Into 6. Here’s What Happened to Query Performance
Benchmarking the impact of fewer, larger files across three SQL workloads
AI Made Data Scientists Faster. Now It’s Expanding the Job
The real shift is bigger than productivity: AI is reshaping ownership, judgment, and the career path of data scientists.
How to Design Architectural Guardrails Around AI Agents
Agent design patterns every data engineer must know
Building Fair Evaluation Sets Is a Combinatorial Problem
Here’s how to solve it exactly.
Guided Merge Sort : An Optimized Sorting that Picks the Best from Ordinary and Multi-Way Merge Sort Algorithms
A solution where the “goto” operator becomes irreplaceable.
How to Make Your Own JEV Model from an Open LLM
Turn a small open-source Qwen LLM into a fast, single-pass text classifier by swapping its language-modeling head
The AI That Learned to Understand Long After It Stopped Trying
A small, strange discovery in machine learning called grokking
How to Catch Data Drift When Every Feature Looks Normal
Detect hidden shifts in feature relationships with adversarial validation and scikit-learn
GraphRAG with TypeSafe Jev: A System One Approach to Scalable Knowledge Graphs
How calibrated decision models can handle high-frequency graph decisions while LLMs remain focused on reasoning, synthesis, and open-ended generation.
Good Architecture Deletes the Signals Your Agent Depends On
Every boundary you draw removes a signal your tooling was relying on. That is a structure problem, not a search problem.
AI Slop Is in Your Training Dataset Now. I Tested Three Ways to Spot It
My AI detectors flagged many genuine reviews, and filtering them made the sentiment model less accurate.
Your LLM Has a Curved Space of Paragraphs
Inside a transformer, token index is a coordinate. Paragraph structure is what turns it into a metric.
10 Things I’m Learning Beyond AI to Become More Technologically Fluent
Part 1: Understanding the technologies shaping our future
Your Model’s MSE Is Lying to You: Part II
Autoregressive rollout and uncertainty propagation. Second in a series on probabilistic forecasting for physical signals.
RAG Isn’t an Agent — I Built the Layer Between Retrieval and Action
RAG retrieves. Agents act. I built both separately, connected them explicitly, and ran the same nine tasks through all three systems.
Jev vs. LLMs: When AI moves from Generation to Decision-making
I tested TypeSafe AI’s Jev on 3,080 classification tasks to see how its accuracy, latency, calibration, and confidence compare with LLMs — and whether it works as a practical decision layer for AI systems.
How to Maximize Your Coding Agent Subscriptions
Get more out of your coding agent subscriptions
Beyond RAGs: Building Actually Truthful AI Harnesses
Retrieval is not evidence. How to build AI that proves its own claims.
Towards Spec-Driven Test Automation: Part 1
Why a green test suite can mean nothing
When the Correct Answer Is Nothing, What Does Your Pipeline Return?
The reliability mechanisms we add to LLM pipelines are often the ones that make them confidently wrong.
I Trained a Tiny Network to Compress Data. It Drew a Pentagon
Reproducing Anthropic’s “Toy Models of Superposition” from scratch in NumPy, with hand-derived gradients and no borrowed numbers.
From Words to Vectors: What Happens in Between?
A Journey through TF-IDF, vector space, and text classification
How GRPO Trains Small Language Models with Verifiable Rewards
The mechanics behind local reasoning experiments with Unsloth and why the reward function matters as much as the model.
How to Make Your First World Model from Scratch
A beginner-friendly guide to building a world model in Python, letting it daydream its way through CartPole, and accurately measuring when the illusion collapses.
Break Your Own RAG Pipeline Before Users Do
A small adversarial test set that catches the retrieval failures your evaluation set never will
Build a Speaker-Recognition App with Claude Code
Learn how to effectively code up an internal tool using Claude code or Codex
4 Ways to Use AI on a PhD Thesis
Finding citations, consolidating code, fact-checking and preparing for the defence
An Introduction to Jev
The AI that makes decisions instead of generating text
A New Kind of Model for AI Decision-Making?
Exploring TypeSafe’s Jev and comparing it with OpenAI on intent classification
GPT-6 Astra Just Hit OpenAI's Highest Cybersecurity Risk Level
Why this matters less for what Astra can do and more for what every other model hasn’t been tested for
Google Offered $10M for a Dying Airline's Data. How Can You Value Yours?
How to put a dollar value on your organization’s operational data
Your AI Assistant Wrote the Code. Who Checked the Defaults?
Five scikit-learn defaults that deserve a closer look before your next model reaches production
GraphRAG: A Practitioner's Guide to 6 Advanced Architectural Patterns
Beyond basic graph retrieval: six production-oriented architectures for combining semantic search, knowledge graphs, and LLM reasoning.
CBAM Paper Walkthrough: The Double-Attention Mechanism
Understanding and implementing CBAM (Convolutional Block Attention Module) from scratch with PyTorch
One Vendor, Four Spellings: How Deterministic Stages Beat Similarity Scores
Deduplicating a 10,000-row supplier list in Python, where the hard part is deciding what a similarity score of 91 means
AI Made Me 5x Faster. It Also Made Me 5x Worse at My Job
One near miss, four months of running agents, and the question almost nobody is asking: what are you supposed to do while the AI writes the code?
Coding Agents Keep Shipping Silent Failures — Here Is How to Catch Them
How to verify your app aligns with your intent without ever reading a line of generated code.
We Pinned Our Model Version to Stay Safe. The Provider Deprecated It Anyway
The recurring cost of production AI is not inference. It is re-qualification: the eval reruns, prompt retuning, and regression testing you owe every time a model changes under you. Here is what that tax actually covers, and how to budget for it before it surprises you.
Starting a Career in Data Science in the Age of AI
How do you set yourself up for a career that will stand the test of time when things are changing so fast?
Multi-Agent Coding Isn’t Enough — Agents Need a Commitment Layer
Multi-agent coding systems don’t necessarily fail because agents can’t communicate. They can fail because important commitments made in conversation have nowhere to live afterward.
Building a Data Lakehouse with DuckDB and DuckLake
Starting with a local Parquet file, then joining it to data stored in the cloud
How I Built a Multi-Agent System for Interrupted Time Series Analysis (ITSA)
What it takes to turn counterfactual analysis into an AI product
Why You Think Like a Bayesian but Were Taught Like a Frequentist
From a chocolate bar with no price tag to a marketing mix model in PyMC, and the 200-year-old integral that stood in between.
When Does Graph RAG Actually Add Value? A Hands-On Experiment
I built four AI retrieval architectures on a laptop and benchmarked them against the same set of documents and questions. Here’s what the results taught me about the trade-offs between plain RAG, graph RAG, and simply putting everything into a frontier model’s context window.
How to Make Linear Regression Survive Outliers
Robust Estimation Series: comparing classical and modern robust estimators through theory, code, and hexperiments
Silent Broadcasting Can Ruin Your Model
PyTorch and TensorFlow tensor broadcasting: how silent shape errors cause difficult-to-debug machine learning bugs
The KV Cache Tax: Why Inference Servers Run Out of Memory Before Compute
A VRAM budget formula for LLM serving, and three optimization strategies mapped to the traffic patterns that trigger the OOM.
The N Squared Pizza Problem
What ordering and not eating a large pizza tells us about ML memory management
Reparameterization Tricks: Variance Reduction by Smarter Gradients
How moving randomness outside the computation graph turns noisy gradient estimators into low-variance, differentiable ones
How to Build Consistent Designs with Claude Code
Keep your apps looking professional with Claude Code design skills
Seizing the Moment: The Hidden Silhouette of Data
How statistical moments connect the mean, the variance, and higher powers of a distribution
How Many Labeled Examples Does a Text Classifier Actually Need? I Measured It
Before reaching for an LLM API on every classification problem, it’s worth knowing what a decades-old baseline can already do with the labeled data you have — and exactly how much more data buys you.