What is it about
Programa completo para convertirte en AI Engineer con Java: construye aplicaciones de IA con Spring AI y LangChain4j, domina Spec-Driven Development para generar código con control y adopta Skill-Driven Development (con KCP) para que el conocimiento de tu equipo componga con la IA.
AI Engineer with Java — one program, three capabilities: build AI applications in Java, steer AI with specifications, and make your team's knowledge compound with skills and KCP.
⚠️ The problem
- Anyone can generate code with AI today, but without control over requirements or architecture
- Every chat starts from zero: you re-explain conventions, domain and past mistakes
- Speed without a safety net → debt, subtle bugs and expensive hallucinations
👉 The problem is not AI.
👉 It is the lack of engineering: specifications, skills and navigable knowledge.
🤖 Block 1 – AI with Java: Spring AI + LangChain4j
Build complete AI applications in Java by integrating LLMs, RAG, agents, memory and tool calling with production-grade architecture.
- Foundations of AI and LLMs applied to Java
- Introduction to Spring AI and LangChain4j
- Configuring Spring AI and connecting to LLM providers
- Prompt engineering (system prompts, few-shot and structured prompts)
- Chatbots with persistent memory
- Embeddings and semantic search with vector databases
- RAG (Retrieval Augmented Generation) in Java
- Tool calling and integration with external APIs
- AI Agents with LangChain4j (tools, memory and orchestration)
- End-to-end AI application architecture (backend + frontend)
📐 Block 2 – Spec-Driven Development with AI
Use specifications as the source of truth and steer AI to implement in Spring Boot and Quarkus without losing control of architecture or tests.
- The problem of AI without engineering: vibe coding vs engineering
- SDD fundamentals and differences vs TDD/BDD — full flow: Spec → Code → Validate
- Constitution: system rules, stack, standards and security
- Writing Specs: user stories, acceptance criteria and edge cases
- Design from the spec: API contracts, data modeling, framework-agnostic architecture
- Multi-framework implementation: Spring Boot and Quarkus
- Development with AI: controlled generation and safe iteration
- Testing from specs: validation against specifications and guardrails
- System evolution: spec-driven changes, safe refactoring and versioning
The specification is the source of truth.
The framework is just one implementation.
🧩 Block 3 – Skill-Driven Development (with KCP)
Treat AI as a system that compounds — not a chat that resets. You are the conductor; AI is the orchestra.
- The 6 pillars: intelligent context, strategic delegation, trust-but-verify, directed synthesis, process discipline and continuous learning
- CLAUDE.md: the codebase DNA (overview, conventions, constraints, test strategy)
- Domain skills: Five Fingerprints to discover your first business skills
- Strategic delegation: fast models for repetitive work; strong models for design
- Verification: round-trips, TDD, QA checklists and catching hallucinations
- KCP (Knowledge Context Protocol): knowledge.yaml with knowledge units (intent, scope, audience, depends_on) so any agent can navigate your project in few tool calls
- Daily SDD workflow on a real backlog task + LEARNINGS.md
🏁 Integrative final project
- A Java AI application built from specifications
- With the project's CLAUDE.md, domain skills and knowledge.yaml (KCP)
- End-to-end validation: specs → code → tests
🎓 Who is this for?
- Backend developers (Java)
- Software architects and tech leads
- Engineers using AI who want to move from prompting to engineering
🧠 Requirements
- Basic to intermediate Java (Spring Boot is a plus)
- REST APIs
- No prior AI or Machine Learning experience required
📦 Includes
- Full project and source code
- Reusable templates:
spec.md,architecture.md,tasks.md,CLAUDE.mdandknowledge.yaml