OpenAI’s agent for software tasks: it explores projects, modifies code, runs checks and presents changes for review. Available in ChatGPT, the editor and the terminal.
A Practical Guide to Artificial Intelligence
Consulting, outreach and specialised training for professionals and companies that want to lead the cognitive era.
Scan to access
Programme Director
Jesus de Pablos
Science communicator, Master’s in AI, Master’s in Cybersecurity and court-appointed technology expert witness.
International speaker, trainer and consultant. He designs AI adoption and training programmes for companies and professionals, with a practical approach that connects automation, cybersecurity and knowledge transfer.
Professional profileJesús de Pablos’s digital assistant
mIA
She helps Jesús research, organise information and prepare content and projects. As head of the AI Agency, she coordinates its specialised agents and brings their contributions together so that Jesús can review the result and make the decisions. She brings experience in creating, evolving and onboarding agentic AI.
Meet mIA and her teamTraining
BootCamp IA
A practical methodology to master the ecosystem of generative AI, agentic AI and workflows.
A programme for professionals and companies that want to apply AI to real tasks: writing better instructions, working with documents, creating content and designing automations and agents. Each session combines content, exercises and Q&A so you can transfer what you learn to your own work.
- 5 sessions of 90 minutes
- Practice and Q&A
- Support material
BootCamp IA
Generative AI
A latest-generation conversational search engine that is transforming digital research by offering synthesised, structured answers with extremely rigorous real-time source citation.
A multimodal assistant for writing and reviewing texts, analysing documents and data, and supporting programming. In a conversation you can provide context, explore ideas and refine answers to create materials tailored to your project.
Google’s ecosystem, with a massive context window and close integration with Workspace, able to process and reason over Drive, Docs and Gmail files in seconds.
xAI’s AI with instant access to the global data stream of X (Twitter). An ideal tool for taking the pulse of current events and getting direct answers without ideological filters.
A leader in logical reasoning and high-quality writing with a deeply human tone. Optimised for safety, code analysis and the processing of long technical documents.
Cognitive AI
Gemini Notebook
A knowledge management system that works with the sources you provide, such as PDFs, web pages and videos. Its limits depend on the plan; citations help you check the answers, which can also contain errors.
Verifiable sources
Audio and video summaries
Document management
Full grounding
Infographics
Compliance
Security and legality: the non-negotiable pillars of corporate AI adoption.
Tool
Anonimiza Tligent
Experimental project
AI browsing
ACT
NORM
Approves the Statute of the Spanish Agency for the Supervision of Artificial Intelligence (AESIA) and regulates its organisation and functions.
AI literacy is addressed separately in Article 4 of Regulation (EU) 2024/1689.
Key concepts to study
Critical skills for professionals.
Prompting and personalisation
Advanced prompt engineering to guide AI towards high-precision business results through the use of specific contexts, tones and roles.
Canvas mode
An interactive workspace for co-creating documents and code in real time, making iterative editing and the refinement of complex content easier.
InPainting
Selective visual editing that lets you modify specific parts of an image while fully respecting the original style, lighting and composition.
Agentic AI
AIs that reason about goals, plan actions and carry out tasks by interacting with external tools autonomously and repeatedly.
Multimodality
The ability of models to understand and generate voice, images and video in a unified way within a single interaction session.
GPTs / Gems
Setting up custom micro-experts with their own system instructions and knowledge files for recurring tasks.
Projects
Workspaces that group documents, instructions and chats under a single shared context for long-term consistency.
Deep Research
Advanced reasoning systems (such as o1 or R1) that take time to "think" before answering complex technical problems.
Learning mode
The ability of platforms to act as personalised tutors, adapting explanations and levels to the user’s progress.
Scheduled tasks
Automating prompts and workflows at set times or frequencies to optimise operational productivity asynchronously.
Critical thinking
The indispensable skill for auditing AI results, ensuring the accuracy of information and keeping ethical judgement in every process.
Specialised AIs
An audio and video transcription service with translation features. Accuracy varies with the language, the quality of the recording and the plan you have.
An assistant that captures live transcripts and generates minutes, action items and sentiment analysis in Zoom, Teams and Meet.
A music composition engine that orchestrates complete pieces with high-fidelity arrangements and vocals from simple text descriptions.
Video creation using photorealistic avatars with voice cloning and multilingual translation that keeps the original timbre.
A creative AI platform to generate, edit and enhance images; check the current features and terms on its website.
Design of high-quality presentations and interactive websites from an initial outline, with no manual layout work.
Google’s creative environment for generating and combining images and video; the old Whisk link leads to Flow.
An experimental Google platform for generating user interfaces (UI) for mobile and web apps from text prompts or reference images such as sketches or wireframes.
A creative mixing environment that turns prompts into interactive canvases with editable images, batch variations and intuitive combinations.
A Google tool for creating mini-apps and automations with AI using only natural language, with no code. It may require a VPN.
Perplexity’s deep-search browser, designed for complex research that requires technical synthesis and rigorous citation.
Open Source & Local AI
Extremely high-performing open models that compete with paid AIs in technical benchmarks for logical reasoning and programming.
A powerful, high-performance foundation model developed by Zhipu AI, focused on logical reasoning and long contexts.
Europe’s pride in open AI. Known for their parameter efficiency and balanced reasoning ability for business solutions.
Local Execution
Run LLMs directly without leaving your computer or private network, with greater control over your data when configured without external services and with appropriate security measures.
A professional interface for discovering and testing Hugging Face models locally in an intuitive way.
OpenAI’s speech recognition engine for top-quality private transcription and translation on your own hardware.
AI-assisted development (Vibe Coding)
A smart code editor (a fork of VS Code) that predicts your next changes, refactors in natural language and understands your entire codebase.
Anthropic’s agentic tool that lives in your terminal. It plans, writes, debugs and runs complex programming tasks autonomously.
A "full stack" platform that turns text descriptions into complete, working web applications (React + Supabase) ready for production.
Agentic AI
Agents that can work step by step with files and tools. Set a goal, limit the permissions and review the result before using it.
Agentic work with Claude: it receives a task and acts on the files and tools you authorise. Anthropic called it Cowork and is integrating its features into Claude.
An open agent from Nous Research with persistent memory and skills. It can connect to messaging apps, schedule tasks and use tools from the desktop or terminal.
An open assistant you can run on your own infrastructure. It coordinates tasks through chat applications, with memory, tools and different models depending on the configuration.
AI Labs
Google’s technical lab for prototyping with Gemini models, managing advanced APIs and generating applications with the Build feature.
A cutting-edge space where Google presents innovative early-stage projects for experimental testing.
LMSYS’s neutral testing ground for evaluating real intelligence through global blind comparisons.
Learning resources
The official technical guide with practical examples for integrating GPT models into business applications.
Official Google training designed for earning professional certifications in AI and data analysis.
Video channels (in Spanish)
A leading Spanish voice focused on generative tools and advanced tutorials for professionals.
A specialist in the practical application of AI in business and productivity automation.
A communicator focused on explaining complex technical concepts in a simple, engaging way.
Carlos Santana. High-level technical outreach on deep learning and fundamental advances in AI.
Training programmes
MiniCamp
60 MINAn essential introductory talk for a first professional contact with AI.
BootCamp IA
INTENSIVEA high-intensity programme of five 90-minute sessions designed for technical and strategic business mastery.
AI Workshop
8 HOURSA fully immersive technical day to solve operational challenges with immediate deployment.
AI Challenges
PRACTICEA platform of challenges designed to apply your knowledge in complex AI environments.
Go to Challenges →Jesus de Pablos
AI communicator and trainer · Court-appointed cybersecurity expert witness
