Blog

When the bug is real but the path isn't: ITScape (CVE-2026-46316) and the case for minimal runtimes

VM isolation at container speed: urunc and the Kubernetes agent-sandbox CRD

Integration of urunc with Kubernetes agent sandbox CRD for VM-based isolation and fast spawn times for AI agent workloads sandboxing

Welcoming Charalampos Mainas

Charalampos Mainas is joining NOFire AI as Member of Technical Staff, Systems. He is one of the creators of urunc, a CNCF sandbox project for fast and lightweight sandboxing, which will be the execution environment that makes truly bounded autonomous action possible.

Design for breach: the agent in production is untrusted by default

The industry just agreed that assume-breach is the operating model. For AI agents acting in production, that means hardware-level isolation and hard controls at the execution boundary, not trust in the model.

The Context & Control Model: The missing piece of the production AI stack

AI agents can't act autonomously in production without a model of how it behaves, and governance that constrains every action they take.

Introducing Workflows: Always-On Agents for Production Reliability

NOFire Workflows are always-on agents that run proactive health checks, generate post-mortems, validate deployments, and respond to production events automatically, with full access to your Production Context Graph.