Block 1. Generative AI Platforms

Module 1: Fundamentals and Overview of Generative AI

✓ The current moment in AI

• What generative AI is: language models, multimodal models, image and video generation
• The four major ecosystems: OpenAI (ChatGPT / GPT‑5.4), Anthropic (Claude Opus 4.6), Google (Gemini 3.1 Pro), Microsoft (Copilot)
• Open‑source models: Llama 4 (Meta), Mistral. When and why they matter
• How LLMs work: tokens, context, probabilities, context windows (non‑technical explanation)
• Key limitations you must know: hallucinations, bias, knowledge cut‑off, confidentiality

✓ AI as a professional competency

• How AI is reshaping the job market: roles gaining value, tasks being automated
• AI as a multiplier: it doesn’t replace professionals—it empowers those who know how to use it
• Including AI in your CV and interviews: how to demonstrate real competence
• The role of human judgment: AI proposes, you decide

✓ Prompting: the cross‑cutting skill

• Anatomy of a good prompt: role + context + task + format + constraints
• Techniques: few‑shot, chain‑of‑thought, delimiters, iterative prompting
• Prompting for professional work: writing, analysis, synthesis, creativity, learning
• Common mistakes and how to avoid them

Module 2: ChatGPT – The Market Standard

✓ ChatGPT in depth

• GPT‑5.4 model (March 2026): advanced reasoning, native computer use, 1M‑token context window
• GPT‑5.4 Thinking: upfront planning, real‑time steerability
• Plans: Free, Plus, Team, Enterprise — what each one includes
• Interface: conversations, history, Projects, memory, custom instructions

✓ Advanced features

• Document analysis: upload PDFs, Word, Excel. Summaries, extraction, comparison
• Data analysis with Code Interpreter: upload CSV/Excel, charts, statistics
• DALL‑E: image generation from text
• GPT‑5.4 Vision: image analysis, screenshots, charts, OCR
• Voice mode: spoken interaction on mobile and desktop
• Deep Research: automated in‑depth research with sources

✓ Custom GPTs

• Creating specialized assistants: instructions, knowledge (files), capabilities
• GPT Store: explore and evaluate community‑built GPTs
• Practical case: creating a Custom GPT to prepare for job interviews

Module 3: Claude – The Deep‑Work Assistant

✓ Claude and the Anthropic ecosystem

• Claude Opus 4.6 (February 2026): Anthropic’s most advanced model, with a 1M‑token window
• Claude Sonnet 4.6 and Haiku 4.5: when to use each one (performance vs. cost vs. speed)
• Claude Desktop: the desktop app with three modes — Chat, Cowork, and Code
• Plans: Pro ($20/month), Max, Team, Enterprise

✓ Claude Chat and Projects

• Analysis of long documents: upload PDFs, contracts, and reports of 100+ pages
• Projects: organize conversations, documents, and context by project
• Artifacts: editable documents, code, and visualizations inside the conversation
• Memory and personalization: how Claude retains context across conversations

✓ Claude Cowork: the desktop agent

• What Cowork is: Claude accesses your computer folders and performs tasks autonomously
• Demonstration: organizing files, generating reports, analyzing documents — Claude works while you do something else
• Cowork Skills: generating professional Word, Excel, PowerPoint, and PDF documents
• Connectors: Google Drive, Gmail, Calendar, Notion, Slack

✓ Claude for Excel and PowerPoint

• Claude add‑ins: AI integrated directly into Excel and PowerPoint
• Shared context (March 2026): one conversation spans both apps without repeating context
• Reusable skills: one‑click workflows shareable across teams

Module 4: Gemini and Google Workspace with AI

✓ Gemini 3.1 Pro and the Google ecosystem

• Gemini 3.1 Pro (February 2026): Google’s most advanced model, with improved reasoning and native multimodality
• Gemini Flash 3.1 and Flash‑Lite: fast, cost‑efficient models
• Gemini in the app: chat, image generation with Nano Banana 2, Deep Research
• Plans: Free, Google AI Pro ($20/month), Google AI Ultra ($25/month)

✓ Gemini integrated into Google Workspace

• Gemini in Google Docs: drafting, rewriting, summarizing
• Gemini in Google Sheets: data analysis, formulas, charts using natural language
• Gemini in Google Slides: presentation generation
• Gemini in Gmail: thread summarization, drafting replies

✓ NotebookLM: research with verifiable sources

• NotebookLM: AI that works only with your documents. Zero hallucinations, automatic citations
• Audio Overviews: turn documents into explanatory podcasts
• Video Overviews, slide decks, infographics, flashcards, and quizzes
• Use case: preparing a civil‑service exam topic or a research project using verified sources

Module 5: Microsoft Copilot – AI in the Office Ecosystem

✓ Microsoft Copilot and M365 Copilot

• Copilot Chat (free): AI‑powered search, content generation, general‑purpose assistant
• Microsoft 365 Copilot (licensed): AI integrated into Word, Excel, PowerPoint, Outlook, and Teams
• Wave 3 (March 2026): Copilot Cowork, in‑app agents, multimodel architecture (GPT‑5.4 + Claude)
• Copilot Studio: no‑code agent creation (introduction — explored in depth in Module 7)

✓ Copilot in action

• Word: drafting, rewriting, summarizing, generating content from other documents.
• Excel: data analysis, formulas, pivot tables, Python in Excel.
• PowerPoint: presentations generated from prompts or documents.
• Outlook and Teams: email thread summaries, drafting replies, meeting notes

✓ Platform comparison: when to use each one

• ChatGPT: best for creative generation, Custom GPTs, and data analysis with Code Interpreter
• Claude: best for extensive document analysis, deep‑work tasks, and autonomous workflows with Cowork
• Gemini: best integration with Google Workspace; NotebookLM for research with verifiable sources
• Copilot: best for organizations already using Microsoft 365
• Selection principle: there is no universally “best” AI — there is a best AI for each task

Block 2. AI in the Real World: Public and Private Sectors

Module 6: AI in the Private Sector and the Public Sector

✓ AI in the Private Sector

• How companies use AI today: marketing, sales, operations, HR, finance, customer service
• Real cases: startups born with AI, large companies integrating it, transformed industries
• Emerging professional roles: prompt engineer, AI product manager, AI ops, agent specialist
• Productivity and competitiveness: companies that do not adopt AI fall behind
• Most in‑demand AI tools in job postings (2026)

✓ AI in the Public Sector

• How public administrations use AI: citizen services, administrative processing, document management, data analysis
• Real cases in Spain and Europe: institutional chatbots, process automation, regulatory analysis
• Civil‑service exams and public employment: how AI is changing the profiles that are needed
• The regulatory Sandbox of AESIA (Spanish Agency for AI Supervision)

✓ Legal and Ethical Framework: What Every Professional Must Know

• EU AI Act: risk categories, obligations, timeline.
• GDPR and AI use: which data must never be entered, anonymization, consent.
• Intellectual property: who owns what AI generates.
• Algorithmic bias: what it is, why it matters, how to detect it.
• Regulatory differences between the public sector (transparency, accountability, ENS) and the private sector (compliance, IP, competitive advantage).
• Responsible use: AI proposes; the professional decides and remains accountable.

Block 3. AI Agents and Agent Teams

Module 7: AI Agents — From Conversation to Action

✓ What is an AI agent?

• The evolution: chatbot → assistant → agent. From answering questions to executing tasks
• Anatomy of an agent: AI model + tools + memory + instructions + autonomy
• Types of agents: conversational (respond), process agents (automate), autonomous agents (decide and act)
• The autonomy spectrum: human‑in‑the‑loop, human‑on‑the‑loop, human‑out‑of‑the‑loop

✓ Agents in each platform

• Custom GPTs (ChatGPT): specialized agents with their own knowledge and tools
• Claude Cowork (Anthropic): a desktop agent that performs tasks on your files
• Claude Code: a development agent that writes, runs, and fixes code autonomously
• Gemini Agents: agents integrated into Google Workspace and Android
• Microsoft Copilot Cowork (Wave 3): an agent operating inside M365 with long‑running tasks
• M365 Agent Builder: create agents from Copilot Chat with no code

✓ Creating agents without programming

• Copilot Studio: Microsoft’s low‑code platform for building agents
• Build an agent in 15 minutes: natural‑language description → functional agent
• Knowledge sources: SharePoint, websites, documents, databases
• Agent Flows: automations the agent executes as tools
• Publish to Teams, the web, Outlook, or as an extension of M365 Copilot

✓ Model Context Protocol (MCP): the connection standard

• What MCP is: an open protocol that allows agents to connect to external tools and data.
• MCP in Claude, Copilot Studio, VS Code, JetBrains: an industry‑wide standard.
• MCP connectors: Gmail, Google Drive, Slack, Notion, GitHub, databases, APIs.
• Why MCP matters for your career: it is the plumbing of agentic AI.

Module 8: Agent Teams — Multi‑Agent Orchestration

✓ From a single agent to an agent team

• Why one agent is not enough: complex tasks require specialization and coordination
• Orchestration patterns: sequential (pipeline), parallel, hierarchical, collaborative
• The concept of roles: one agent researches, another analyzes, another drafts, another reviews
• Agent‑to‑agent communication: A2A protocol, delegation, shared outputs

✓ Agent teams in practice

• Claude Cowork with sub‑agents: Claude breaks down a complex task into subtasks and coordinates multiple sub‑agents in parallel
• Microsoft Agent 365 + Copilot Studio: agents calling other agents, A2A, multi‑system orchestration
• Squad (GitHub): pre‑built agent teams for development (lead, frontend, backend, tester)
• Reference frameworks: CrewAI, LangGraph, Anthropic Agent SDK, Google ADK — high‑level overview

✓ Governance of agents and teams

• Observability: knowing what each agent does, when, why, and with what outcome
• Human‑in‑the‑loop: when and how a human intervenes in an agent team
• Security and permissions: what each agent is allowed to do, data access, destructive actions
• Audit and compliance: traceability of decisions made by agents
• Governance differences: public sector (mandatory auditing, transparency) vs. private sector (efficiency, IP)

✓ The future of work with agents

• The professional as an “agent boss”: supervising agent teams instead of executing tasks
• New roles: agent designer, workflow orchestrator, AI supervisor
• How to position yourself professionally: agent‑related skills are the new competitive advantage