Top all-time · #15339
Multikernel Technologies
Company blog · multikernel.iosystemsaisecurity
Multikernel Technologies is the company blog at multikernel.io, writing about systems, ai and security. Since 2025, 5 different Hacker News readers have linked it 10 times, most often “Per-Tool Sandboxing for AI Agents: Why One Sandbox Is Not Enough”. It is still writing: the last post was on 2026-10-25, and you can follow it by feed.
Too few sampled comments (3) to say how often HN recommends it.?
Why HN links it
HN commenters link the multikernel.io site and the related mklinux project, which run several Linux kernels at once on bare metal. Comments are bare links or neutral references, with no judgment of the writing.
Summarised by Claude from a sample of 3 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Multikernel turns servers you own and VMs you rent into a private cloud: bare-metal performance, kernel-level isolation, no hypervisor, no virtualization tax.
Most-linked posts?
- Per-Tool Sandboxing for AI Agents: Why One Sandbox Is Not Enough2 people · 2 mentions · first linked 2026-03-26
- /2 people · 2 mentions · first linked 2025-09-19
What Hacker News says?
Interestingly the author has a startup revolving around this technology. Their webpage has some info: https://multikernel.io/
Comment linking /
on “Kernel: Introduce Multikernel Architecture Support”Submitted AI Agent Sandboxes Got Security Wrong as a story · 3 points
HN discussionhttps://github.com/multikernel/linux https://multikernel.io/
Comment linking /
on “Mklinux Multikernel: Run several Linux kernels at once, bare metal, no VT-x”Submitted One Agent Sandbox Is Not Enough as a story · 1 points
HN discussionI'm curious if you've looked at a fundamentally different solution - namely, mklinux[1]? You not only skip the entire KVM/Firecracker VMM/Kata Shim/Guest Linux/Kata Agent stack which achieves your goal of reducing the attack surface, using mklinux also mitigated entire classes of vulnerabilities such as buffer overflows/out-of-bounds I/O in the virtio stack, speculative execution side-channels, SLAT bugs etc. [1] ht…
Comment linking /technology.html
on “Secure VMs for Kubernetes: Hardening Kata Containers”Submitted One Pipe, Two Sandboxes, Zero Prompt Injection as a story · 1 points
HN discussionSubmitted Per-Tool Sandboxing for AI Agents: Why One Sandbox Is Not Enough as a story · 1 points
HN discussionSubmitted Sandlock vs. Containers: 25% Faster as a story · 1 points
HN discussionSubmitted COW Fork: Zero-Copy Sandbox Cloning for AI Agents as a story · 1 points
HN discussionSubmitted Processes Are All You Need for AI Sandboxing as a story · 1 points
HN discussion