# LearnIA — Applied AI Engineering Training

> LearnIA (learn-prompting.fr) is a free, open-access AI education platform offering 10 interactive guides and 150+ articles on prompt engineering, RAG, AI agents, multimodal AI, ethics, and context engineering. All content is available in English and French, with guides accessible in 7 languages.

## Site Structure

- Homepage: https://learn-prompting.fr
- Blog (EN): https://learn-prompting.fr/blog
- Blog (FR): https://learn-prompting.fr/fr/blog
- Guides Hub: https://learn-prompting.fr/guides

## Interactive Guides (10 modules — all free, no account required)

- [Prompt Engineering Basics](https://learn-prompting.fr/guides/prompt-engineering-basics): AI fundamentals, the 4 pillars of prompting (R.C.T.F), first exercises. Beginner, ~1.5h.
- [LLM Fundamentals](https://learn-prompting.fr/guides/llm-fundamentals): How tokens work, 5-component prompt structure, zero/one/few-shot techniques. Beginner, ~1h.
- [Structured AI Outputs](https://learn-prompting.fr/guides/structured-ai-outputs): JSON extraction, schema validation, CV extractor workshop. Beginner, ~1h10.
- [Chain-of-Thought Reasoning](https://learn-prompting.fr/guides/chain-of-thought-reasoning): CoT, Self-Consistency, Tree-of-Thought, reasoning arena. Intermediate, ~1h10.
- [Prompt Orchestration](https://learn-prompting.fr/guides/prompt-orchestration): Chaining, routing, map-reduce patterns for complex workflows. Intermediate, ~50min.
- [RAG — Retrieval-Augmented Generation](https://learn-prompting.fr/guides/rag-retrieval-augmented-generation): Vector search, chunking, embeddings, mini-RAG workshop. Intermediate, ~50min.
- [AI Agents](https://learn-prompting.fr/guides/ai-agents): ReAct method, tool calling, guardrails, agent safety. Advanced, ~1h15.
- [Multimodal Creative AI](https://learn-prompting.fr/guides/multimodal-creative-ai): Diffusion models, visual prompt anatomy, creative workshop. Advanced, ~1h.
- [AI Ethics & Safety](https://learn-prompting.fr/guides/ai-ethics-safety): Hallucinations, bias detection, red teaming, GDPR compliance. Advanced, ~1h10.
- [Context Engineering](https://learn-prompting.fr/guides/context-engineering): 4 pillars, lost-in-the-middle, prompt caching, MCP protocol. Expert, ~1h20.

## Key Topics Covered

### Prompt Engineering
- Zero-shot, one-shot, and few-shot prompting techniques
- The 5 components of an effective prompt (Role, Context, Instruction, Format, Constraints)
- Iterative refinement and prompt version control
- Building a personal prompt library

### RAG (Retrieval-Augmented Generation)
- How RAG works: retrieve → augment → generate
- Chunking strategies (fixed, recursive, semantic, contextual)
- Embedding models, vector databases (FAISS, Pinecone, Weaviate)
- Building a mini RAG system from scratch

### AI Agents
- The ReAct (Reasoning + Acting) framework
- Thought → Action → Observation loops
- Tool calling and function integration
- Safety guardrails and human-in-the-loop patterns

### Advanced Reasoning
- Chain-of-Thought (CoT) prompting
- Self-Consistency voting across multiple reasoning paths
- Tree-of-Thought (ToT) exploration
- Reasoning arenas for comparing approaches

### Context Engineering
- The 4 pillars: Write, Select, Compress, Isolate
- Lost-in-the-middle problem and positional bias
- Prompt caching for cost and latency optimization
- Model Context Protocol (MCP) for tool integration

### Ethics & Safety
- AI hallucination detection and mitigation
- Bias identification and fairness testing
- Red teaming techniques for AI systems
- GDPR / EU AI Act compliance for AI applications
- Building an AI ethics charter

## Key Interactive Features

- 3D vector space visualization (React Three Fiber)
- Prompt injection security lab (Learn → Attack → Defend)
- Reasoning technique arena (5 techniques × 5 challenges)
- RAG pipeline simulator (chunk → embed → retrieve → generate)
- Agent simulation workbench (real-time Thought-Action-Observation traces)
- Context window simulator (drag-and-drop placement with scoring)
- Mermaid diagrams for process visualization

## Content Format

- Articles in Markdown/MDX with interactive components
- Interactive quizzes, exercises, prompt builders, JSON validators
- Code examples in Python and JavaScript
- Bilingual: English (primary) + French
- Guides available in 7 languages: EN, FR, DE, ES, IT, RU, TA

## Blog Categories

- Prompt Engineering (30+)
- AI Agents (30+)
- AI Coding (45+)
- RAG (12+)
- Advanced Reasoning (15+)
- Structured Outputs (10+)
- Context Engineering (5+)
- AI Ethics (18+)
- Multimodal AI (12+)

## Sitemap

https://learn-prompting.fr/sitemap.xml

## Contact

- Website: https://learn-prompting.fr
- Languages: English, French, German, Spanish, Italian, Russian, Tamil
- [GPT-5.2 Codex Deep Dive](https://learn-prompting.fr/blog/gpt-5-2-codex-deep-dive)

## Contact

- Website: https://learn-prompting.fr
- Company: LearnIA by LK Digital (https://lkdigital.fr)