Operations | Monitoring | ITSM | DevOps | Cloud

If you run Claude through LiteLLM, some of that spend is probably going uncounted – and you can’t see it, precisely because the data isn’t there. Routing through a gateway is messier than it looks: LiteLLM alone can carry Claude several ways – the OpenAI-compatible endpoint, and the Anthropic pass-through proxy that the native SDK and Claude Code use – and each path describes the same call differently.

In the first experiment, I wanted a baseline: if an AI coding agent gets the same production signal a human would get, can it fix bugs in a codebase it has never seen? Yes, but only when I gave it better context. With only an alert, the agent passed 51% of the runtime tests. When I added captured traffic, the actual request and response for the failing call, it climbed to 77%. This post is the second pass.

We rebuilt docs.cloudzero.com from the ground up. The result is a documentation site organized around what you’re trying to accomplish rather than feature names and jargon.

Why do 40% of alerts received by security teams today go completely uninvestigated? It’s not due to a lack of concern but instead caused by shortening attack windows and compounded by overwhelming tech sprawl. Today’s security teams are operating in a threat landscape defined by escalating attacks, tighter budgets and mounting alert fatigue. Organizations process an average of 960 security alerts per day, and large enterprises handle more than 3,000 daily alerts across roughly 30 tools.

New research reveals why Australian businesses are entering the new financial year with bigger AI budgets and the same unsolved problem. One in three Australian businesses exceeded their AI budget last year. Yet, half of them plan to increase AI spending again this year. Yet the behaviour that caused those budget overruns remains largely unaddressed.

AI agents are nondeterministic, multi-step, and opaque. When one fails in production, "the model said something weird" is the cheapest, most useless line in your incident postmortem. To debug agents the way they actually run, you need telemetry that captures all of it, in order, with enough context to reconstruct what happened. The OpenTelemetry GenAI Semantic Conventions give you a vendor-neutral way to do exactly that.