Insights
Latest thinking
Technical deep-dives on building AI systems that work in production. Opinionated, evidence-backed, no fluff.
Why Most Teams Get Agentic Observability Wrong (And What to Instrument Instead)
Teams building AI agents reach for the same observability tools they use for microservices — and end up blind to the failures that actually matter. The problem isn't a lack of data; it's instrumenting the wrong layer entirely.