Operations | Monitoring | ITSM | DevOps | Cloud

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.

IT’S HERE it’s here it’s here it’s here!!!! The second edition of Observability Engineering is available for download, and since Honeycomb is the sponsor, you can now download it from our website (the dead tree version will take another month). This is a strange time to be writing a book.

A few weeks ago I wrote about a customer’s refund request that stopped halfway through at 11:47 p.m. on a Tuesday night. That post walked through the 40 minutes it took to work out what happened when an agentic application had a problem: a tool retried against a rate-limited payments API, the error responses filled up the context window, and the agent gave up. The whole reason we built Agent Timeline was to turn that 40 minutes into five. To reduce MTTR. To solve the problem and get back to sleep.

Observability has always been important, and much like any core capability in your business, the value needs to be understood. For years, the value of observability was predictable. It was uptime, error rates, MTTR, and likely tool consolidation. That was enough to be able to show progress. These are foundational, tablestakes metrics—and they still matter, but they aren’t enough.

This is the fourth installment in the Graviton retrospective series we've been writing since 2021. The methodology is the same one I always reach for: hold the workload constant, run both generations on the same Kubernetes namespace concurrently, and let the per-pod numbers speak.

The OpenTelemetry Collector is usually deployed as a long-running process: a sidecar, a DaemonSet, an EC2 instance, a docker container on my computer. It sits there listening for telemetry. That's fine when I want to send telemetry all day, but not when telemetry is rare. Like right now, when I have an agent defined on AgentCore, and it runs a few times a week maybe. Or my website that hardly sees any traffic. Can I run the OpenTelemetry Collector as a Lambda function?

When a measure becomes a target, it ceases to be a good measure. Charles Goodhart, 1975 You’ve probably read this quote in relation to any number of things over the years. People complaining about arbitrary metrics like PRs merged, lines of code produced, and now, token usage. But is the era of tokenmaxxing over before it even began? The rise of token leaderboards to the death of token leaderboards at companies like Amazon seem to have taken place in less than three months!

You'd think that working at an observability company means everyone knows exactly where to find everything in the data. It doesn't. Especially not on the support team. We're the ones who get the tickets. We're in the telemetry every day trying to figure out what went wrong for a customer, and we do that by pointing Honeycomb at itself. Here's how that actually works, and how it's changed.

We just wrapped O11yCon 2026, and this year's conversations hit differently. Agent-based software development is here, now. It's no longer an optional choice, and everybody is struggling to understand what their agents are doing and how to make them cost less and perform better. Over the course of fifteen talks, we saw clearly that the old assumptions on how and who (or what) writes our software has been upended. Here are some highlights. We'll have videos available in the near future.

Last week, we launched a major update to Canvas, our investigation workspace. The new Canvas has evolved from an AI co-pilot you chat with to a place where your whole team, human and agent, can work the same problem on the same surface. Auto-investigations begin the moment a trigger, SLO, or anomaly fires. Custom skills encode your team's runbooks so every agent investigates with your team's expertise built in.