Monitoring, tracing, and understanding what your applications are doing in production.
Observability is more than dashboards: it's about understanding your systems deeply. These articles cover OpenTelemetry, distributed tracing, metrics, logging strategies, and the tools that help you debug production issues before your users notice.
My obsession with observability started when I began building AI agents with Genkit and had no idea where the hits were coming from when things broke down. My articles come from that real-world pain: instrumenting production agents and cloud infrastructure to turn blind debugging into clear, actionable traces.
Guía práctica sobre observabilidad para agentes de IA, OpenTelemetry, trazas, spans y la arquitectura en 3 capas para depurar sistemas de IA sin dar golpes a ciegas.