How to use it in favour of thinking, not against it.
AI models do not remember previous conversations, but they can process enormous amounts of text in a single session. Understanding how this space works changes how you work with them.
Read →
Language models can write for you, but the real risk is not that they make mistakes — it is that we all start to sound the same. Using AI well in writing requires knowing exactly where it adds value and where it gets in the way.
Read →
Not all AI models work the same way. Reasoning models spend explicit time thinking before they answer. Understanding when and why to use them can make the difference between a mediocre response and a genuinely useful one.
Read →
A good prompt isn't always enough. Sometimes you need to train the model on your own data. But fine-tuning has a real cost and isn't always the right answer.
Read →
How Claude Code works, what sets it apart from its competitors and why it's changing the way developers write software.
Read →
What Google Antigravity 2.0 is, how its agent system works and why Google is betting on a radically different approach.
Read →
What OpenAI Codex offers, how it integrates with ChatGPT and its open-source CLI, and who it makes the most sense for in 2026.
Read →
Using AI tools is not the same as getting value from them. Before adding more to your workflow, it's worth knowing whether the ones you already use are working.
Read →
Large language models know a lot, but they know nothing about your documents. Retrieval-augmented generation solves that problem by connecting AI to your own information.
Read →
A new AI tool promising to transform your workflow appears every week. A framework for deciding which ones are worth exploring and which to ignore.
Read →
Reasoning models don't just generate text — they deliberate before responding. Understanding what that process involves changes how you should use them.
Read →
AI models no longer process only text. They understand images, audio, and video. This expands what you can do with them — and changes how you need to think to use them well.
Read → Language models are not neutral: they can reinforce what you already believe or help you see more clearly. The difference lies entirely in how you use them.
Read →
Not every task deserves your attention. AI can handle the routine so you can focus on what matters.
Read → Running a language model on your own hardware offers privacy and control that cloud services cannot guarantee. But there are real trade-offs worth understanding before you commit.
Read →An AI agent does not just answer questions — it makes decisions and executes tasks autonomously. This changes what it means to work with artificial intelligence.
Read →
Neither magic nor apocalypse. Artificial intelligence is a family of statistical techniques that learn patterns from data. Understanding this changes everything else.
Read → What actually happens when an AI model 'learns': data, parameters, loss functions and gradient descent. No formulas, just logic.
Read → Supervised, unsupervised, reinforcement learning, generative. The full taxonomy with real examples for each type, so you can stop using 'AI' as if it were one single thing.
Read →
From the first perceptrons to transformers. Why AI failed twice before it worked, and what changed in 2017 to make everything different.
Read →
Large language models do not understand: they predict. The difference has enormous practical implications for knowing when to trust them and when to verify.
Read → A token is not a word. Context is not memory. Understanding these distinctions prevents half the mistakes people make when working with language models.
Read → RLHF, instruction tuning and values. How you go from a model that predicts text to one that follows instructions, responds with judgement, and avoids causing harm.
Read → Hallucinations are not technical glitches you can patch out — they are consequences of the prediction mechanism. Warning signs and verification strategies for using AI without nasty surprises.
Read →
Role, context, task and format. The four elements of a prompt that works, with comparative examples to understand the difference between vague and precise instructions.
Read →
Three techniques that multiply output quality. When to use each one and how to combine them for complex problems that a simple prompt does not solve.
Read → How to ask for JSON, tables, lists or specific formats. Stop AI from improvising structure when you need precision, and learn to connect AI outputs with real systems.
Read → The first response is rarely the best. How to build on what AI produces to get closer to the desired result without starting over every time.
Read →
The AI ecosystem in 2025–2026. Which tool is best for which task, and why mastering one well rather than five badly makes the real difference in productivity.
Read → AI does not write for you — it accelerates the parts that consume the most time. How to use it at each stage of the writing process without losing your voice or your judgement.
Read → How to process large volumes of information with AI without swallowing the errors. The verification method that saves you from surprises when what the model produces is not what it seemed.
Read → From a list of tools to a system. How to integrate AI into real work without the tool becoming another source of friction or distraction.
Read → Stable Diffusion, Midjourney, DALL·E. The logic of reverse noise that converts text into image, and why visual prompting works differently to text prompting.
Read → Transcription, voice synthesis, cloning and video generation. What works, what fails and what raises questions that go beyond the technology.
Read → Copilot, Claude, Cursor. How AI is changing software development for experts and non-experts alike, and where the real limits are that are worth knowing.
Read → Retrieval-Augmented Generation. How to make a model respond about your internal documentation without retraining it or sharing sensitive data with an external service.
Read → A chatbot responds. An agent decides, acts and observes the result. The architectural difference that makes it possible, the risks it introduces and why understanding it matters.
Read → How an agent uses external tools, maintains context across actions and chains reasoning to complete tasks that a simple chatbot cannot tackle.
Read → Agents are fragile in real environments. Cases where they add genuine value, common design errors and how to supervise them without losing the benefit of autonomy.
Read → Biases in AI are not accidents — they reflect training data and design decisions. How to identify them, when they matter and what to do about them.
Read → What providers do with your conversations. How to work with sensitive information without compromising your own or third parties' data without realising it.
Read → AI does not replace people — it replaces tasks. How to build a productive relationship with the technology, what skills will remain irreplaceable and what is worth learning now.
Read → Artificial intelligence is advancing fast. But there are capabilities it fundamentally lacks — not for want of power, but due to structural limitations. It pays to know which ones.
Read → A chatbot answers questions. An AI agent makes decisions, executes actions, and chains tasks autonomously. The difference is not one of degree — it is one of nature.
Read → AI adoption in the workplace has reached a tipping point in Spain. Understanding the gap between those using it and those who aren't is more urgent than it was a year ago.
Read → AI models tend to agree with you. If you do not know how to ask them not to, you get validation instead of usefulness.
Read → Not to decide for you. To help you see what you are missing before you decide.
Read → Every time you type in an AI chat, you're sharing information. Understanding what happens to that information isn't paranoia — it's digital literacy.
Read → AI can dramatically accelerate the process of learning about a new topic. But doing it well means understanding its limitations before trusting its answers.
Read → The quality of what you get from an AI depends almost entirely on the quality of what you ask. Here's the difference.
Read → You don't need to automate everything or become a programmer. There's a set of repetitive, cognitively expensive tasks that AI handles better than you.
Read → Artificial intelligence does extraordinary things. It also fails in ways worth understanding before you depend on it.
Read → Language models invent data with total confidence. Learning to detect it is a basic skill for anyone who uses them at work.
Read → A real workflow, no tricks, for integrating AI as a reflection tool.
Read →