How Are You Isolating AI Agents From Each Other in Kubernetes?
For those running AI agents in production on Kubernetes, what isolation boundary are you actually using between agents? Most deployments seem to place them in ordinary pods, which…
microVMs
What level of isolation are you using for AI agents in Kubernetes?
For teams running AI agents in production on Kubernetes, what security boundary do you put between agents and between an agent and the host? Regular pods provide namespaces,…
What’s the Best Architecture for a Developer-Friendly AI Agent Sandbox?
I'm exploring an infrastructure layer for running AI agents inside isolated, virtualized environments. The intended flow is: AI agent → sandbox API or SDK → Firecracker microVM →…
What Hard Problems Do Lambda MicroVMs Actually Unlock?
With Lambda MicroVMs becoming available, what genuinely difficult or previously impractical workloads can they handle? I'm especially interested in common problems where strong isolation, fast startup, pause/resume, statefulness,…
Are microVMs useful beyond AI agents and serverless workloads?
Over the past few months, several companies have released sandboxing tools for AI agents based on microVM technology. I currently run nginx inside a virtual machine on my…