---
title: "Artificial intelligence concepts"
seo_title: "Artificial intelligence concepts explained with technical rigor"
description: "Answer-first guides to LLMs, Transformers, agents, prompt injection, reasoning and AI evaluation, connected to learning series and engineering notes."
keywords: "artificial intelligence concepts, LLM, Transformer, AI agents, prompt injection, AI reasoning, model evaluation"
hide:
  - toc
  - navigation
  - footer
---

<div class="s5-landing">

<section class="s5-page-intro">
  <div class="s5-eyebrow">Concepts</div>
  <h1>A direct entry point to the ideas behind modern AI.</h1>
  <p>These guides answer the essential question first and then open up the mechanism. Each concept links to the chapters and engineering notes where it is applied in practice.</p>
</section>

<section class="s5-section" aria-labelledby="core-concepts">
  <div class="s5-section-head"><h2 id="core-concepts">Technical foundations</h2></div>
  <div class="s5-simple-list">
    <a class="s5-list-row" href="/en/temas/llms/">
      <span class="s5-list-row__n">01</span>
      <span class="s5-list-row__title">What is an LLM?</span>
      <span class="s5-list-row__desc">Tokens, pretraining, generation, alignment and the limits of large language models.</span>
      <span class="s5-list-row__meta">Guide</span>
    </a>
    <a class="s5-list-row" href="/en/temas/transformer/">
      <span class="s5-list-row__n">02</span>
      <span class="s5-list-row__title">How the Transformer works</span>
      <span class="s5-list-row__desc">Attention, representations, residual blocks and the architecture that enabled modern models to scale.</span>
      <span class="s5-list-row__meta">Architecture</span>
    </a>
    <a class="s5-list-row" href="/en/temas/razonamiento/">
      <span class="s5-list-row__n">03</span>
      <span class="s5-list-row__title">Reasoning in LLMs</span>
      <span class="s5-list-row__desc">Chain of thought, search, verifiers, test-time compute and why more steps do not guarantee a better answer.</span>
      <span class="s5-list-row__meta">Inference</span>
    </a>
    <a class="s5-list-row" href="/en/temas/evaluacion-modelos/">
      <span class="s5-list-row__n">04</span>
      <span class="s5-list-row__title">Evaluating AI models</span>
      <span class="s5-list-row__desc">How to move from an isolated benchmark to a valid evaluation of the system and product.</span>
      <span class="s5-list-row__meta">Measurement</span>
    </a>
    <a class="s5-list-row" href="/en/temas/agentes-ia/">
      <span class="s5-list-row__n">05</span>
      <span class="s5-list-row__title">What is an AI agent?</span>
      <span class="s5-list-row__desc">Chatbot vs workflow vs agent, tools, memory, state, evaluation, permissions and when agency is actually useful.</span>
      <span class="s5-list-row__meta">Systems</span>
    </a>
    <a class="s5-list-row" href="/en/temas/prompt-injection/">
      <span class="s5-list-row__n">06</span>
      <span class="s5-list-row__title">What is prompt injection?</span>
      <span class="s5-list-row__desc">Why data and instructions can compete inside context, and which architectural boundaries reduce the path from untrusted content to action.</span>
      <span class="s5-list-row__meta">Security</span>
    </a>
  </div>
</section>

<section class="s5-section">
  <div class="s5-note-feature">
    <div>
      <div class="s5-eyebrow">Next level</div>
      <h2>From concept to system.</h2>
      <p>The learning series develop each idea progressively. The engineering notes show what changes once latency, state, tools, audio and human interaction become part of the system.</p>
      <a class="s5-text-link" href="/en/series/">Explore the series →</a><br />
      <a class="s5-text-link" href="/en/articulos-tecnicos/">Open the engineering notes →</a>
    </div>
    <div class="s5-note-feature__meta">Definition<br />Mechanism<br />Evidence<br />Production</div>
  </div>
</section>

</div>