PTENES
Track 1

🌱 AI Fundamentals

Build a solid foundation in artificial intelligence. Understand the current landscape, master prompt engineering, and learn about context and automation to transform your career and projects.

5
Modules
30
Topics
~6h
Duration
Basic
Level
1.1 ~30 min

🌊 The AI Tsunami

What it is: Large language models (LLMs) are neural networks trained on vast text datasets to understand, generate, and manipulate human language. Examples include GPT, Claude, LLaMA, and Gemini.
Why learn: LLMs are the foundation of the current generative AI revolution. Understanding how they work helps you use these tools more efficiently and identify real opportunities to apply them in your work.
Key concepts: Tokens and tokenization, context window, pre-training and fine-tuning, attention mechanism (Transformers), inference and temperature.
What it is: Exponential growth describes the pace of AI's progress—significantly more capable models emerge every 6-12 months, costs fall, and access increases.
Why learn: Understanding the exponential nature of change prevents surprises. Those who underestimate how quickly things change fall behind. Those who understand it can position themselves strategically.
Key concepts: Moore's Law applied to AI, scaling laws, benchmark improvements, model commoditization, adoption speed compared with previous technologies.
What it is: The history of AI begins in the 1950s with Alan Turing and the Turing Test, passes through winters and revivals, and reaches the era of Transformers (2017) and today’s generative AI.
Why learn: Knowing the history helps you understand how we got to this point and avoid repeating past mistakes. It also provides context for evaluating the technology's promises and limitations.
Key concepts: Turing test, AI winters, classical machine learning, deep learning, Transformers, GPT and the generative era, historical milestones (AlphaGo, ChatGPT).
What it is: Technological convergence refers to the simultaneous combination of advances in AI, cloud computing, big data, IoT, and connectivity, which together create an unprecedented landscape for transformation.
Why learn: AI doesn’t exist in isolation. Understanding the converging forces helps you identify innovation opportunities where technologies intersect.
Key concepts: Cloud computing, big data, APIs, 5G and connectivity, platform economy, digital ecosystems, network effects.
What it is: AI is transforming nearly every sector of the economy. Finance uses AI for fraud detection, healthcare for diagnostic imaging, education for personalized learning, and so on.
Why learn: Regardless of your field, AI will have a direct impact. Understanding sector-specific impacts helps you anticipate changes and prepare professionally.
Key concepts: FinTech and AI, HealthTech, EdTech, AgriTech, LegalTech, industrial automation, smart logistics, predictive marketing.
What it is: Practical strategies for positioning yourself advantageously in the face of the AI wave: adopt a continuous learning mindset, experiment with tools, and build real projects.
Why learn: Knowing change is coming isn't enough—you need to act. This topic turns knowledge into concrete action for your career.
Key concepts: Growth mindset, continuous learning, project portfolio, networking in AI, practical first steps, an experimental mindset.
View Full
1.2 ~35 min

💼 Digital Transformation and the Job Market

What it is: The Fourth Industrial Revolution (Industry 4.0) is the fusion of physical, digital, and biological technologies—AI, robotics, IoT, biotechnology, and quantum computing are converging to transform every aspect of society.
Why learn: We are living through this revolution now. Unlike previous ones, it is advancing exponentially and affecting every profession at the same time.
Key concepts: Industry 4.0 (Klaus Schwab), physical-digital convergence, cyber-physical systems, technological disruption, speed vs. breadth of impact.
What it is: According to the World Economic Forum, by 2027, 9 million jobs will be eliminated by automation, while 11 million new positions will be created—resulting in a net gain but requiring massive reskilling.
Why learn: Knowing which jobs are at risk and which are emerging lets you plan your career strategically and invest in the right skills.
Key concepts: Future of Jobs Report (WEF), reskilling and upskilling, jobs at risk, emerging jobs, career transition, skills gap.
What it is: Research indicates that 70% of the skills required in the job market will change significantly by 2030. Technical skills will have increasingly shorter lifespans, while human skills will become more important.
Why learn: Investing in the right skills is the best protection against professional obsolescence. Understanding this dynamic is essential for any career plan.
Key concepts: Half-life of skills, hard skills vs. soft skills, critical thinking, creativity, emotional intelligence, adaptability, digital literacy.
What it is: Some sectors are being transformed more rapidly: finance (algorithmic trading, risk analysis), manufacturing (robotics and predictive maintenance), healthcare (AI-assisted diagnosis), and services (automated customer support).
Why learn: If you work in one of these sectors, the need to adapt is even more urgent. If not, understanding these cases can help you anticipate what’s coming to your field.
Key concepts: Financial automation, Industry 4.0 in manufacturing, AI in medical diagnosis, chatbots and customer service, retail transformation, impact on the legal sector.
What it is: The T-shaped model describes professionals with depth in one area of expertise (the vertical bar of the T) and breadth of knowledge in complementary areas (the horizontal bar), including AI.
Why learn: In today's market, being only a specialist or only a generalist isn't enough. The T-shaped + AI model is the formula for becoming indispensable.
Key concepts: T-shaped professional, Pi-shaped professional, specialization vs. generalization, unique combination of skills, AI as a multiplier of skills.
What it is: AI literacy is the ability to understand, use, and critically evaluate AI tools. It doesn't mean knowing how to code, but knowing how to interact effectively with intelligent systems.
Why learn: Just as digital literacy was essential in the 2000s, AI literacy is the defining skill of this decade. Those who do not develop it will be at a significant disadvantage.
Key concepts: AI Literacy, advanced digital literacy, critical use of AI, algorithmic bias, AI ethics, digital citizenship in the age of AI.
View Full
1.3 ~40 min

💬 Prompt Engineering

What it is: A prompt-building technique that starts with general instructions and progressively refines them until they reach the level of detail needed to get precise responses from AI.
Why learn: Most users get mediocre results because they ask vague questions. Mastering progressive specificity dramatically improves the quality of AI responses.
Key concepts: Context, persona, output format, constraints, examples, tone and style, prompt iteration, progressive refinement.
What it is: A technique that assigns a role or persona to AI—such as "You are a digital marketing expert with 20 years of experience"—activating more specific and relevant knowledge patterns.
Why learn: Role-playing is one of the most powerful and simple prompt engineering techniques. It significantly improves the quality, depth, and relevance of responses.
Key concepts: System prompts, personas, simulated expertise, professional tone, multiple perspectives, simulated debate among experts.
What it is: Chain-of-Thought (CoT) is a technique that asks AI to “think step by step,” spelling out its reasoning before reaching a conclusion. This significantly improves accuracy on complex tasks.
Why learn: For problems involving logic, math, analysis, or complex decisions, CoT can be the difference between a wrong answer and a correct one.
Key concepts: Explicit reasoning, zero-shot CoT, few-shot CoT, Tree-of-Thought, reasoning verification, logical decomposition.
What it is: Few-Shot Learning is the technique of providing examples within the prompt so the AI understands exactly the desired input and output pattern, with no additional training required.
Why learn: When you need specific formats, consistent styles, or complex response patterns, few-shot is the most reliable technique for getting consistent results.
Key concepts: Zero-shot, one-shot, few-shot, positive and negative examples, example-based formatting, output consistency, reusable templates.
What it is: A technique that breaks complex problems into smaller, manageable subtasks, each solved separately by AI, with the results combined at the end.
Why learn: LLMs have limitations with very complex tasks in a single prompt. Breaking tasks down improves quality and makes it possible to solve problems that seemed impossible.
Key concepts: Divide and conquer, sub-prompts, prompt pipelines, orchestration, sequencing, intermediate validation.
What it is: An overview of the main generative AI platforms available: ChatGPT (OpenAI), Claude (Anthropic), Gemini (Google), Copilot (Microsoft), and specialized tools for images, video, and audio.
Why learn: Each platform has different strengths and weaknesses. Knowing how to choose the right tool for each task multiplies your productivity and the quality of your results.
Key concepts: ChatGPT vs. Claude vs. Gemini, open-source models, APIs, playgrounds, costs and limits, multimodal tools (DALL-E, Midjourney, Suno).
View Full
1.4 ~35 min

🧩 Context Engineering

What it is: Context engineering is the discipline of structuring and providing relevant information to AI in an optimized way. Its 4 pillars are: context selection, compression, ordering, and dynamic updating.
Why learn: The quality of an AI’s response depends directly on the quality of the context you provide. Context engineering is the next level beyond prompt engineering.
Key concepts: Context window, contextual relevance, information compression, context window management, meta-prompting, system instructions.
What it is: RAG is an architecture that combines searches across external databases with AI text generation, enabling answers grounded in current, specific data.
Why learn: RAG solves two of the biggest problems with LLMs: hallucinations and outdated knowledge. It's the most widely used technique in enterprise AI applications.
Key concepts: Retrieval, indexing, chunking, knowledge bases, vector databases, semantic search, grounding responses, hallucination reduction.
What it is: Embeddings are numerical representations (vectors) of text that capture semantic meaning. Texts with similar meanings are close together in vector space, enabling similarity searches.
Why learn: Embeddings are the foundation of RAG systems, recommendation engines, and intelligent search. Understanding the concept is essential for building advanced AI applications.
Key concepts: Vectors, vector space, cosine similarity, embedding models, Pinecone, Weaviate, FAISS, semantic vs. lexical search.
What it is: Knowledge graphs organize information as networks of entities and relationships, allowing AI to understand complex connections between concepts and make more sophisticated inferences.
Why learn: For applications that require reasoning about complex relationships (e.g., compliance, scientific research), knowledge graphs complement embeddings and significantly improve quality.
Key concepts: Nodes and edges, triples (subject-predicate-object), Neo4j, GraphRAG, ontologies, relational inference, Knowledge Graph + LLM.
What it is: Function Calling allows LLMs to invoke external functions — query APIs, access databases, perform calculations — transforming AI from a text generator into an agent that performs real actions.
Why learn: Function Calling is the bridge between conversational AI and real automation. It is the foundation of autonomous agents and assistants that actually do things for you.
Key concepts: Tool use, function definition, structured parameters, APIs, webhooks, autonomous agents, tool orchestration.
What it is: Memory systems for AI replicate concepts from human memory: short-term (conversation context), long-term (persistent knowledge), and episodic (history of past interactions).
Why learn: Memory is what turns a simple chatbot into a truly useful assistant that knows you, your preferences, and your history over time.
Key concepts: Short-term memory (buffer), long-term memory (vector store), episodic memory, conversation summarization, user profiles, progressive personalization.
View Full
1.5 ~35 min

🤖 Process Automation

What it is: Intelligent automation combines traditional rule-based automation with artificial intelligence, enabling the automation of processes involving judgment, natural language interpretation, and decision-making.
Why learn: Intelligent automation is the biggest productivity multiplier available today. It lets one person do the work of an entire team in certain tasks.
Key concepts: RPA vs. intelligent automation, IPA (Intelligent Process Automation), automated decision-making, document processing, intelligent classification.
What it is: A methodology for documenting and analyzing existing processes, identifying which steps are repetitive, time-consuming, or prone to errors—and therefore ideal candidates for automation.
Why learn: Automating the wrong process wastes resources. Mapping ensures you automate what truly generates impact and returns.
Key concepts: Process mining, flowcharts, bottlenecks, cycle time, prioritization matrix, quick wins, complexity vs. impact.
What it is: No-code/low-code platforms that let you create visual automations by dragging and connecting blocks, with no programming required. They integrate hundreds of apps and services.
Why learn: No-code tools democratize automation. Any professional can automate their processes without relying on developers, saving hours each week.
Key concepts: Zapier (ease of use), Make.com (flexibility), n8n (open source), triggers and actions, webhooks, API integration, automation templates.
What it is: Combining RPA (Robotic Process Automation) with AI creates software robots that don’t just follow scripts, but also understand documents, make decisions, and learn over time.
Why learn: RPA + AI is one of the fastest-growing areas in enterprise technology. Professionals who master this combination are highly valued in the market.
Key concepts: UiPath, Automation Anywhere, document understanding, intelligent OCR, process discovery, attended vs unattended bots, hyperautomation.
What it is: Framework for calculating the return on investment of automation projects, considering implementation costs, time savings, error reduction, and scalability gains.
Why learn: To justify investments in automation (personal or business), you need to demonstrate concrete returns. Knowing how to measure ROI turns automation from a "hobby" into a strategy.
Key concepts: TCO (Total Cost of Ownership), payback period, hours saved, error reduction, customer satisfaction, productivity metrics, business case.
What it is: A phased implementation methodology: start with a small pilot project, validate results, make adjustments, and expand gradually, minimizing risks and maximizing learning.
Why learn: Most automation projects fail because they try to do everything at once. Gradual implementation is the approach with the highest proven success rate.
Key concepts: Automation MVP, pilot project, validation metrics, change management, scaling, center of excellence (CoE), automation governance.
View Full