Operations | Monitoring | ITSM | DevOps | Cloud

During the Toronto KCD (Kubernetes Community Days), I attended an insightful talk on AI resource optimization that highlighted a staggering Gartner study: “AI infrastructure is adding $401 billion in new spending this year alone. Yet, real-world audits tell a much darker story, revealing that average GPU utilization in the enterprise is stuck at a dismal 5%”. While many people in the audience were shocked by that number, the data didn’t come as a surprise to us.

Your Dynamic Host Configuration Protocol (DHCP) server already knows what’s connected to your network. The problem is that DHCP data rarely stays aligned with the rest of your infrastructure systems. Instead, it becomes fragmented across Windows servers, branch offices, spreadsheets, and disconnected operational tools. Lease data ages, assignments go untracked, and teams lose confidence in their network inventory.

For Platform Engineering teams, the goal has always been clear: build a secure, scalable internal developer platform that reduces cognitive load and accelerates time-to-market. Yet, a massive obstacle often remains hidden in plain sight: the mainframe. While your distributed teams are shipping cloud-native microservices multiple times a day, your core backend mainframe applications frequently remain locked in an isolated silo, lagging behind on slow monthly or quarterly cadences.

I'm happy to share some really exiting updates from the Flyway product teams this past month.

Claude Mythos is now available to the public through Claude Fable 5, released June 9, 2026. Claude Fable 5 pricing is $10 per million input tokens and $50 per million output tokens, exactly 2x Claude Opus 4.8 ($5/$25). Claude Mythos 5 (the restricted Project Glasswing version) has identical pricing. Prompt caching cuts input spend by 90%. Batch API pricing is $5/$25 (50% off). In April 2026, Anthropic announced a model it said was too dangerous to release.

AI spend has no ceiling. An engineer can burn $5,000 in an hour, and a team that spins up an agent on Friday can loop it on a bad prompt all weekend. You find out when the bill lands: the money is already gone, the damage pieced back together from logs. Cloud spend had a natural limit. Tokens don’t. Now you see it as it happens. Connect a source and the calls stream in within seconds. Within minutes they’re broken out by model, provider, agent, and user.