Operations | Monitoring | ITSM | DevOps | Cloud

Walk into a network operations center this year and the useful AI is not running the place. It is doing three specific jobs, and doing them well: cutting an alert storm down to the one incident that matters, pointing at the likely cause, and deciding what deserves a human’s attention first. That is where AI in NetOps pays for itself right now. The part worth noticing is that all three jobs lean on the same thing.

Model routing promises to cut your AI agent spend by offloading routine tasks to cheaper models like Claude Haiku while reserving frontier models like Claude Sonnet for complex reasoning. In the right configuration, routing strategies can reduce inference costs by 40–85%. But if you implement routing incorrectly in a multi-turn agent, you can end up paying more than if you’d never routed at all. Here’s why and how to fix it.

Coding agents changed implementation economics faster than they changed confidence. They let us produce more code, more quickly, but they did not make reviewers any better at understanding system-wide consequences. In our Kubernetes automation stack, that gap became impossible to ignore once AI started generating meaningful amounts of controller code.

I was at the dentist the other day, getting ready for my appointment. The waiting room was pompously decorated. Each chair seemed to be from a different, expensive Danish designer. As I realize I’m about to get charged through the nose, I get a notification from my beloved Mail app. ** ding ** Screenshot of GitHub email notification It’s a new Pull Request on GitHub. This one is different though. I have no idea where it came from!

In December 2025, an AI coding agent at AWS suddenly decided to delete and rebuild an entire production environment, causing a 13-hour service disruption and a PR headache for Amazon. As rapid adoption of AI leads to more high-profile, revenue-impacting incidents, resilience has moved from a technical concern to a board-level financial risk.

Your engineers are running Claude Code every day, and every prompt burns tokens you’re paying for. Until now, that spend was hard to see. It either sat invisible or landed in an untagged bucket you couldn’t break down. Claude Code already emits detailed telemetry for every interaction, so the data existed. You just had nowhere to send it that would turn it into a cost.

Uber ran out of its entire 2026 AI budget by April. This didn’t happen because AI technology failed, but because the company had no way to connect what it spent to what it got. The COO described it on an earnings call: “It’s very hard to draw a line” between AI usage and consumer product outcomes. And with that one sentence, we have the CFO problem of 2026.

GPT-5.6 pricing runs across three tiers, per million tokens. Sol costs $5 input / $30 output. Terra costs $2.50 / $15. Luna costs $1 / $6. All three share a 1.05 million token context window. The twist nobody priced in: OpenAI’s own system card admits Sol sometimes takes action nobody approved, then reports the job as done. For finance teams, that behavior is a governance issue worth understanding before engineering routes production traffic to it.