HN Showcase
Back to browse
GitHub Repository

Millisecond microVM sandbox forking for AI agents on Kubernetes. Firecracker VMs that restore from memory snapshots in milliseconds, fork a running VM into N copies, and persist durable, versioned workspaces. Self-hostable, declarative CRDs.

3 starsGo

Mitos – N-way live copy-on-write fork of running Firecracker microVMs

by stubbi·Jun 20, 2026·3 points·0 comments
Visit ProjectView on HN

AI Analysis

●●●BangerWizardryBig BrainNiche Gem

Copy-on-write VM forking at 27ms beats Firecracker's native snapshot restore by an order of magnitude.

Strengths
  • •Verifiable ~27ms P50 warm-claim activate latency with public benchmark scripts
  • •Kubernetes CRD integration means existing infra teams can adopt without workflow changes
  • •Self-hosted option addresses data sovereignty concerns that cloud-only alternatives ignore
Weaknesses
  • •Extremely narrow audience—only matters if you're running AI agents on K8s already
  • •Competes with gVisor and direct Firecracker usage which are more established
Target Audience

Teams running AI agents on Kubernetes who need fast isolated environments

Similar To

Firecracker · gVisor · Kata Containers

Similar Projects

AI/ML●●●Banger

NervOS – Sandbox for AI Agents Using Firecracker MicroVMs

Hardware-isolated VM sandbox for Claude, 2-second boot, no Docker complexity.

Zero to OneWizardrySolve My Problem
ashishch111
103mo ago