# Jacar > Notas de producción sobre tecnología, DevOps e IA desde Madrid. Technical blog by Javier Cañete, CEO of Jacar Systems. Covers DevOps, applied AI, Kubernetes, observability, cybersecurity, and product engineering. Content is published in Spanish and English. ## Key pages - [Sobre / About](https://jacar.es/sobre/): About the site and its author. - [Javier Cañete](https://jacar.es/sobre/): Author profile and posts. ## Recent posts - [Qué es Docker Agent, el plugin para crear agentes de IA con YAML](https://jacar.es/que-es-docker-agent/): Docker Agent crea y ejecuta agentes de IA desde la línea de comandos de Docker: YAML declarativo, herramientas MCP y orquestación de varios agentes. - [What Docker Agent is, the CLI plugin for building AI agents in YAML](https://jacar.es/en/what-is-docker-agent/): Docker Agent is a Docker CLI plugin for building and running AI agents from declarative YAML, with MCP tools and multi-agent orchestration on any provider. - [Qué es Open GSD, el flujo Git-Ship-Done para agentes de programación](https://jacar.es/que-es-open-gsd/): Open GSD (Git. Ship. Done.) reúne herramientas libres para dirigir agentes de programación con un bucle de planificar, ejecutar y verificar sin perder contexto. - [What Open GSD is, the Git-Ship-Done loop for coding agents](https://jacar.es/en/what-is-open-gsd/): Open GSD (Git. Ship. Done.) is an open-source toolkit that steers coding agents through a plan, execute and verify loop so they do not lose context mid-project. - [What is a vector embedding and what is it used for](https://jacar.es/en/what-is-a-vector-embedding/): A vector embedding converts text into numbers that preserve semantic meaning: learn how they work, what they are used for, and how to pick a model. - [Qué es un embedding vectorial y para qué sirve](https://jacar.es/que-es-un-embedding-vectorial/): Un embedding vectorial convierte texto en números que preservan su significado semántico: aprende cómo funcionan, para qué sirven y cómo elegir el modelo. - [RAG with Postgres and pgvector in production: from PoC to SLO](https://jacar.es/en/rag-with-postgres-and-pgvector-in-production-from-poc-to-slo/): Deploy RAG in production with Postgres and pgvector: HNSW indexing, reranking, recall evaluation, and measurable SLOs. - [RAG con Postgres y pgvector en producción: del PoC al SLO](https://jacar.es/rag-postgres-pgvector-produccion/): Desplegar RAG en producción con Postgres y pgvector: indexado HNSW, reranking, evaluación de recall y SLOs medibles. - [NIS2 in Spain: a technical translation of 2026 obligations](https://jacar.es/en/nis2-in-spain-a-technical-translation-of-2026-obligations/): NIS2 Spain: technical obligations, incident notification windows (24h/72h/1m) and supply chain security enforceable since Oct 2024 under EU directive. - [NIS2 en España: traducción técnica de las obligaciones para 2026](https://jacar.es/nis2-espana-traduccion-tecnica/): NIS2 España: obligaciones técnicas, ventanas de notificación y cadena de suministro que ya son exigibles en 2026 según la directiva UE. - [EU AI Act 2026: a technical checklist for Spanish CTOs](https://jacar.es/en/eu-ai-act-2026-a-technical-checklist-for-spanish-ctos/): EU AI Act checklist 2026: system classification, obligations, transparency and penalties. Download technical template to audit your AI systems. - [EU AI Act 2026: checklist técnica para CTOs españoles](https://jacar.es/eu-ai-act-checklist-cto-espana/): EU AI Act checklist 2026: clasificación, obligaciones, transparencia y sanciones. Descarga plantilla técnica para auditoría de sistemas IA. - [Agent observability with OpenTelemetry GenAI semconv in 2026](https://jacar.es/en/agent-observability-with-opentelemetry-genai-semconv-in-2026/): Agent observability with OpenTelemetry GenAI semconv. Instrument LLMs, tools and MCP servers in Python. Query traces in Grafana Tempo. Stable spec 2026. - [Observabilidad de agentes con OpenTelemetry GenAI semconv en 2026](https://jacar.es/otel-genai-observabilidad-agentes/): Observabilidad de agentes con OpenTelemetry GenAI semconv. Instrumenta LLMs, tools y MCP servers en Python. Grafana Tempo para consultar trazas. - [How to install and tune oMLX on M5 Max 128 GB](https://jacar.es/en/how-to-install-and-tune-omlx-on-m5-max-128-gb/): oMLX on Mac: local LLM server with TurboQuant 3.5-bit and Qwen 35B. Claude Code integration, configuration guide and benchmarks for Mac M5 Max 128GB. - [Cómo instalar oMLX en M5 Max 128 GB y exprimirlo al máximo](https://jacar.es/omlx-llm-local-mac/): oMLX en Mac: servidor LLM local con TurboQuant 3,5-bit y Qwen 35B. Integración Claude Code, guía de configuración y benchmarks para Mac M5 Max 128GB. - [Multi-agent systems: LangGraph vs CrewAI vs Autogen in 2026](https://jacar.es/en/multi-agent-systems-langgraph-vs-crewai-vs-autogen-in-2026/): Multi-agent systems: LangGraph vs CrewAI vs Autogen. Mental models, when to use each, real case study. - [Sistemas multi-agente: LangGraph vs CrewAI vs Autogen en 2026](https://jacar.es/multi-agente-langgraph-crewai-autogen/): Sistemas multi-agente: LangGraph vs CrewAI vs Autogen. Modelos mentales, cuándo usar cada uno, caso práctico. - [What PegaProx adds over the Proxmox VE 9 GUI](https://jacar.es/en/what-pegaprox-adds-over-the-proxmox-ve-9-gui/): PegaProx vs Proxmox VE 9: AGPL-3.0 multi-cluster panel with live migration, OIDC, CVE scanner and one-click hardening. - [Qué añade PegaProx sobre la GUI de Proxmox VE 9](https://jacar.es/pegaprox-vs-proxmox-ve-9-gui/): PegaProx vs Proxmox VE 9: panel multi-clúster AGPL-3.0 con migración entre clústeres, OIDC, escáner CVE y hardening en un clic.