{"slug": "firecracker", "name": "Firecracker", "category": "infra", "category_name": "Infrastructure and DevOps", "description": "Open-source virtual machine monitor from AWS for running lightweight, secure microVMs on Linux.", "recommenders": 12, "commenters": 20, "warners": 0, "score": 12.707, "rising": 1.38, "first": "2026-01-01", "latest": "2026-10-05", "why": "HN commenters like Firecracker for VM-level isolation in self-hosted CI runners and AI agent sandboxes, and several call it easy to work with.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 11, "neutral": 5, "compare": 4, "warn": 0}, "spark": [0, 0, 1, 2, 0, 1, 1, 1, 1, 3, 0, 1], "listed": true, "loved": ["Agent sandboxes", "CI runners"], "caveat": "Complex operations", "short_quote": "we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "icon": true, "domain": null, "github_repo": "github.com/firecracker-microvm/firecracker", "official_url": "https://firecracker-microvm.github.io/", "pricing": "open source", "aliases": ["Firecracker"], "verdict": [{"text": "HN commenters like Firecracker for VM-level isolation", "cites": [{"id": 789373106, "item_id": 49335819, "url": "https://news.ycombinator.com/item?id=49335819", "by": "srcreigh", "date": "2026-08-17", "stance": "recommend", "quote": "It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "excerpt": "Yeah. MicroVMs make hosting runners pretty straightforward. I had codex make a forgejo runner controller which pulls forgejo actions and creates kubernetes jobs for them. It was a bit of a pain to configure firecracker with k3s. It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "context": "", "use_case": "microVM CI runners", "switched_from": "", "story": "Ask HN: Alternatives to GitHub"}, {"id": 775706944, "item_id": 48481684, "url": "https://news.ycombinator.com/item?id=48481684", "by": "_pdp_", "date": "2026-06-10", "stance": "recommend", "quote": "we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "excerpt": "Haha, we don't use Kubernetes. It is a single large server with plenty of space for runners. It is as simple and boring as it gets. The only fun part is that we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "context": "runs CI runners on a single large server", "use_case": "isolated CI runners", "switched_from": "", "story": "Surprise Pay $1000"}, {"id": 743366976, "item_id": 46460436, "url": "https://news.ycombinator.com/item?id=46460436", "by": "ammmir", "date": "2026-01-01", "stance": "neutral", "quote": "first with Firecracker microVM", "excerpt": "I've built the same thing twice, first with Firecracker microVM, and second time using containers (gVisor). While the microVM route is more secure, it's more complicated and ops are tricky, but you can do some cool things to optimize startup time like when I was working on a function as a service platform, and to reduce TTFB, I trapped the `listen()` call, sent a VSOCK message to the VMM to trigger a freeze, snapshot", "context": "built sandbox platforms", "use_case": "sandboxing / function as a service", "switched_from": "", "story": "Ask HN: What is the best microVMs for AI agents?"}]}, {"text": " in self-hosted CI runners and AI agent sandboxes,", "cites": [{"id": 751180097, "item_id": 46948756, "url": "https://news.ycombinator.com/item?id=46948756", "by": "adamcharnock", "date": "2026-02-09", "stance": "recommend", "quote": "We've created a Firecracker-based runner, which will run CI jobs in Firecracker VMs.", "excerpt": "We've migrated to Forgejo over the last couple of weeks. We position ourselves[0] as an alternative to the big cloud providers, so it seemed very silly that a critical piece of our own infrastructure could be taken out by a GitHub or Azure outage. It has been a pretty smooth process. Although we have done a couple of pieces of custom development: 1) We've created a Firecracker-based runner, which will run CI jobs in ", "context": "runs a cloud provider alternative company", "use_case": "CI job VMs", "switched_from": "", "story": "It's not you; GitHub is down again"}, {"id": 775616562, "item_id": 48476035, "url": "https://news.ycombinator.com/item?id=48476035", "by": "_pdp_", "date": "2026-06-10", "stance": "neutral", "quote": "a Go service that orchestrates firecracker vms", "excerpt": "I am sure Blacksmith is the easy path if you don't want to do any extra work and I am also sure their solution will scale much better as this is what they do full-time. However, I can share what we did to ease on our GitHub Actions bills if it helps. Effectively we have our own runners hooked so that a job is scheduled, a runner picks it up and goes with it. We still use GitHub Actions but our monthly bill is now fla", "context": "", "use_case": "pre-warmed CI runner VMs", "switched_from": "", "story": "Surprise Pay $1000"}, {"id": 786594899, "item_id": 49162181, "url": "https://news.ycombinator.com/item?id=49162181", "by": "dregitsky", "date": "2026-08-03", "stance": "recommend", "quote": "Firecracker VMs + support for \"forking\" the box were our key requirements.", "excerpt": "For boxes.dev we use E2B - they had everything we needed at the time. I'd love a standardized API / marketplace, but more just to try out new providers with differentiated features or pricing - we probably wouldn't be switching often / using multiple (unless customers asked us for more options). There are so many sandbox providers popping up! Our use case is pretty specific though - one persistent machine that you se", "context": "builds boxes.dev", "use_case": "VM isolation for agent boxes", "switched_from": "", "story": "Devtools must be open source"}, {"id": 755121698, "item_id": 47195106, "url": "https://news.ycombinator.com/item?id=47195106", "by": "buremba", "date": "2026-02-28", "stance": "recommend", "quote": "use proper VMs like Kata Containers and Firecrackers.", "excerpt": "My take is that agents should only take actions that you can recover from by default. You can gradually give it more permission and build guardrails such as extra LLM auditing, time boxed whitelisted domains etc. That's what I'm experimenting with https://github.com/lobu-ai/lobu 1. Don't let it send emails from your personal account, only let it draft email and share the link with you. 2. Use incremental snapshots an", "context": "", "use_case": "isolating agents in production", "switched_from": "", "story": "Don't trust AI agents"}]}, {"text": " and several call it easy to work with.", "cites": [{"id": 775706944, "item_id": 48481684, "url": "https://news.ycombinator.com/item?id=48481684", "by": "_pdp_", "date": "2026-06-10", "stance": "recommend", "quote": "we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "excerpt": "Haha, we don't use Kubernetes. It is a single large server with plenty of space for runners. It is as simple and boring as it gets. The only fun part is that we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "context": "runs CI runners on a single large server", "use_case": "isolated CI runners", "switched_from": "", "story": "Surprise Pay $1000"}, {"id": 799520833, "item_id": 49970052, "url": "https://news.ycombinator.com/item?id=49970052", "by": "johnsmith1840", "date": "2026-10-05", "stance": "recommend", "quote": "Firecracker and gvisor are nice systems not horrible to use", "excerpt": "I've deploy gvisor, done basic test of firecracker and an honest attempt at production kata. Firecracker and gvisor are nice systems not horrible to use, gvisor isn't quite the same security level though. Kata is HARD to make. The technical know how to make that in production is awe inspiring. I wanted to use it but it was so complicated to integrate into a cluster I literally just gave up and mirrored raw VMs into t", "context": "", "use_case": "microVM isolation", "switched_from": "", "story": "Linux containers in 500 lines of code (2016)"}]}, {"text": " The most common caveat is that it is harder to set up and operate than containers,", "cites": [{"id": 743366976, "item_id": 46460436, "url": "https://news.ycombinator.com/item?id=46460436", "by": "ammmir", "date": "2026-01-01", "stance": "neutral", "quote": "first with Firecracker microVM", "excerpt": "I've built the same thing twice, first with Firecracker microVM, and second time using containers (gVisor). While the microVM route is more secure, it's more complicated and ops are tricky, but you can do some cool things to optimize startup time like when I was working on a function as a service platform, and to reduce TTFB, I trapped the `listen()` call, sent a VSOCK message to the VMM to trigger a freeze, snapshot", "context": "built sandbox platforms", "use_case": "sandboxing / function as a service", "switched_from": "", "story": "Ask HN: What is the best microVMs for AI agents?"}, {"id": 762627267, "item_id": 47664204, "url": "https://news.ycombinator.com/item?id=47664204", "by": "benswerd", "date": "2026-04-06", "stance": "recommend", "quote": "For hardware virtualized machines it much harder but you can do it via:", "excerpt": "Self hosting can be doable for constant small/medium size workloads You can handroll a lot with: https://github.com/nestybox/sysbox?tab=readme-ov-file https://gvisor.dev https://github.com/containers/bubblewrap?tab=readme-ov-file For hardware virtualized machines it much harder but you can do it via: https://github.com/firecracker-microvm/firecracker/ https://github.com/cloud-hypervisor/cloud-hypervisor Freestyle/oth", "context": "", "use_case": "hardware-virtualized sandboxes", "switched_from": "", "story": "Launch HN: Freestyle: Sandboxes for AI Coding Agents"}]}, {"text": " especially alongside Kubernetes.", "cites": [{"id": 789373106, "item_id": 49335819, "url": "https://news.ycombinator.com/item?id=49335819", "by": "srcreigh", "date": "2026-08-17", "stance": "recommend", "quote": "It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "excerpt": "Yeah. MicroVMs make hosting runners pretty straightforward. I had codex make a forgejo runner controller which pulls forgejo actions and creates kubernetes jobs for them. It was a bit of a pain to configure firecracker with k3s. It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "context": "", "use_case": "microVM CI runners", "switched_from": "", "story": "Ask HN: Alternatives to GitHub"}]}], "praise": [{"label": "Agent sandboxes", "summary": "Commenters recommend it or build on it to isolate AI agents, and note that sandbox providers such as E2B use it.", "quotes": [{"id": 786594899, "item_id": 49162181, "url": "https://news.ycombinator.com/item?id=49162181", "by": "dregitsky", "date": "2026-08-03", "stance": "recommend", "quote": "Firecracker VMs + support for \"forking\" the box were our key requirements.", "excerpt": "For boxes.dev we use E2B - they had everything we needed at the time. I'd love a standardized API / marketplace, but more just to try out new providers with differentiated features or pricing - we probably wouldn't be switching often / using multiple (unless customers asked us for more options). There are so many sandbox providers popping up! Our use case is pretty specific though - one persistent machine that you se", "context": "builds boxes.dev", "use_case": "VM isolation for agent boxes", "switched_from": "", "story": "Devtools must be open source"}, {"id": 755121698, "item_id": 47195106, "url": "https://news.ycombinator.com/item?id=47195106", "by": "buremba", "date": "2026-02-28", "stance": "recommend", "quote": "use proper VMs like Kata Containers and Firecrackers.", "excerpt": "My take is that agents should only take actions that you can recover from by default. You can gradually give it more permission and build guardrails such as extra LLM auditing, time boxed whitelisted domains etc. That's what I'm experimenting with https://github.com/lobu-ai/lobu 1. Don't let it send emails from your personal account, only let it draft email and share the link with you. 2. Use incremental snapshots an", "context": "", "use_case": "isolating agents in production", "switched_from": "", "story": "Don't trust AI agents"}, {"id": 793690320, "item_id": 49605645, "url": "https://news.ycombinator.com/item?id=49605645", "by": "RohanAdwankar", "date": "2026-09-08", "stance": "neutral", "quote": "they use firecracker with memory in git", "excerpt": "I took a look inside the VMs that a couple of the agent platforms are using. TLDR is they use firecracker with memory in git. I'm a big fan of the memory in git part it would be cool to see how the git history looks after I use it for a longer period of time. This is my first blog, I'm trying to document things publically as I learn, next I'll look more closely at the providers they are using under the hood (eg. e2b)", "context": "", "use_case": "VMs for mobile agent platforms", "switched_from": "", "story": "The VMs Powering Mobile Agents (Instinct, Claude Code)"}]}, {"label": "CI runners", "summary": "Teams run CI jobs in Firecracker VMs and say microVMs make hosting runners much easier.", "quotes": [{"id": 789373106, "item_id": 49335819, "url": "https://news.ycombinator.com/item?id=49335819", "by": "srcreigh", "date": "2026-08-17", "stance": "recommend", "quote": "It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "excerpt": "Yeah. MicroVMs make hosting runners pretty straightforward. I had codex make a forgejo runner controller which pulls forgejo actions and creates kubernetes jobs for them. It was a bit of a pain to configure firecracker with k3s. It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "context": "", "use_case": "microVM CI runners", "switched_from": "", "story": "Ask HN: Alternatives to GitHub"}, {"id": 751180097, "item_id": 46948756, "url": "https://news.ycombinator.com/item?id=46948756", "by": "adamcharnock", "date": "2026-02-09", "stance": "recommend", "quote": "We've created a Firecracker-based runner, which will run CI jobs in Firecracker VMs.", "excerpt": "We've migrated to Forgejo over the last couple of weeks. We position ourselves[0] as an alternative to the big cloud providers, so it seemed very silly that a critical piece of our own infrastructure could be taken out by a GitHub or Azure outage. It has been a pretty smooth process. Although we have done a couple of pieces of custom development: 1) We've created a Firecracker-based runner, which will run CI jobs in ", "context": "runs a cloud provider alternative company", "use_case": "CI job VMs", "switched_from": "", "story": "It's not you; GitHub is down again"}, {"id": 775616562, "item_id": 48476035, "url": "https://news.ycombinator.com/item?id=48476035", "by": "_pdp_", "date": "2026-06-10", "stance": "neutral", "quote": "a Go service that orchestrates firecracker vms", "excerpt": "I am sure Blacksmith is the easy path if you don't want to do any extra work and I am also sure their solution will scale much better as this is what they do full-time. However, I can share what we did to ease on our GitHub Actions bills if it helps. Effectively we have our own runners hooked so that a job is scheduled, a runner picks it up and goes with it. We still use GitHub Actions but our monthly bill is now fla", "context": "", "use_case": "pre-warmed CI runner VMs", "switched_from": "", "story": "Surprise Pay $1000"}]}, {"label": "Strong isolation", "summary": "Commenters say it provides real VM isolation that is more secure than container-based options.", "quotes": [{"id": 775706944, "item_id": 48481684, "url": "https://news.ycombinator.com/item?id=48481684", "by": "_pdp_", "date": "2026-06-10", "stance": "recommend", "quote": "we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "excerpt": "Haha, we don't use Kubernetes. It is a single large server with plenty of space for runners. It is as simple and boring as it gets. The only fun part is that we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "context": "runs CI runners on a single large server", "use_case": "isolated CI runners", "switched_from": "", "story": "Surprise Pay $1000"}, {"id": 743366976, "item_id": 46460436, "url": "https://news.ycombinator.com/item?id=46460436", "by": "ammmir", "date": "2026-01-01", "stance": "neutral", "quote": "first with Firecracker microVM", "excerpt": "I've built the same thing twice, first with Firecracker microVM, and second time using containers (gVisor). While the microVM route is more secure, it's more complicated and ops are tricky, but you can do some cool things to optimize startup time like when I was working on a function as a service platform, and to reduce TTFB, I trapped the `listen()` call, sent a VSOCK message to the VMM to trigger a freeze, snapshot", "context": "built sandbox platforms", "use_case": "sandboxing / function as a service", "switched_from": "", "story": "Ask HN: What is the best microVMs for AI agents?"}, {"id": 799520833, "item_id": 49970052, "url": "https://news.ycombinator.com/item?id=49970052", "by": "johnsmith1840", "date": "2026-10-05", "stance": "recommend", "quote": "Firecracker and gvisor are nice systems not horrible to use", "excerpt": "I've deploy gvisor, done basic test of firecracker and an honest attempt at production kata. Firecracker and gvisor are nice systems not horrible to use, gvisor isn't quite the same security level though. Kata is HARD to make. The technical know how to make that in production is awe inspiring. I wanted to use it but it was so complicated to integrate into a cluster I literally just gave up and mirrored raw VMs into t", "context": "", "use_case": "microVM isolation", "switched_from": "", "story": "Linux containers in 500 lines of code (2016)"}]}, {"label": "Docker replacement", "summary": "Some teams chose Firecracker over Docker, and one ML engineer wished they had switched sooner.", "quotes": [{"id": 772659520, "item_id": 48291220, "url": "https://news.ycombinator.com/item?id=48291220", "by": "spwa4", "date": "2026-05-27", "stance": "recommend", "quote": "I should have switched to running docker containers in Firecracker looooong ago.", "excerpt": "The thing I find infuriating about docker is once you try to use it for serious machine learning, you absolutely require mounting your data. On a small scale we're talking s3 mounts, for large scale it's ceph mounts or (please don't) NFS. Oh and it must be an actual mount because speed. Or to put it differently: if you don't use kernel mounts your AI engineers need to design their training loop 10x better. It can wor", "context": "ML infrastructure", "use_case": "running ML containers", "switched_from": "Docker", "story": "Building an AWS Lambda-Like Runtime with Firecracker MicroVMs"}, {"id": 743370032, "item_id": 46460627, "url": "https://news.ycombinator.com/item?id=46460627", "by": "zfoong", "date": "2026-01-01", "stance": "recommend", "quote": "We were thinking about using Docker and eventually settled on Firecracker.", "excerpt": "Thank you for your guidance! We were thinking about using Docker and eventually settled on Firecracker. Also, an interesting project you got there. If you are interested, would it be possible to invite you over to our project Discord? Would love to hear more of your experience.", "context": "", "use_case": "microVMs for AI agents", "switched_from": "Docker", "story": "Ask HN: What is the best microVMs for AI agents?"}]}, {"label": "Easy to use", "summary": "Some users describe it as nice and very easy to work with.", "quotes": [{"id": 775706944, "item_id": 48481684, "url": "https://news.ycombinator.com/item?id=48481684", "by": "_pdp_", "date": "2026-06-10", "stance": "recommend", "quote": "we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "excerpt": "Haha, we don't use Kubernetes. It is a single large server with plenty of space for runners. It is as simple and boring as it gets. The only fun part is that we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "context": "runs CI runners on a single large server", "use_case": "isolated CI runners", "switched_from": "", "story": "Surprise Pay $1000"}, {"id": 799520833, "item_id": 49970052, "url": "https://news.ycombinator.com/item?id=49970052", "by": "johnsmith1840", "date": "2026-10-05", "stance": "recommend", "quote": "Firecracker and gvisor are nice systems not horrible to use", "excerpt": "I've deploy gvisor, done basic test of firecracker and an honest attempt at production kata. Firecracker and gvisor are nice systems not horrible to use, gvisor isn't quite the same security level though. Kata is HARD to make. The technical know how to make that in production is awe inspiring. I wanted to use it but it was so complicated to integrate into a cluster I literally just gave up and mirrored raw VMs into t", "context": "", "use_case": "microVM isolation", "switched_from": "", "story": "Linux containers in 500 lines of code (2016)"}]}, {"label": "Snapshots and forking", "summary": "Commenters use VM snapshotting and cloning for fast startup and for forking full machines.", "quotes": [{"id": 743366976, "item_id": 46460436, "url": "https://news.ycombinator.com/item?id=46460436", "by": "ammmir", "date": "2026-01-01", "stance": "neutral", "quote": "first with Firecracker microVM", "excerpt": "I've built the same thing twice, first with Firecracker microVM, and second time using containers (gVisor). While the microVM route is more secure, it's more complicated and ops are tricky, but you can do some cool things to optimize startup time like when I was working on a function as a service platform, and to reduce TTFB, I trapped the `listen()` call, sent a VSOCK message to the VMM to trigger a freeze, snapshot", "context": "built sandbox platforms", "use_case": "sandboxing / function as a service", "switched_from": "", "story": "Ask HN: What is the best microVMs for AI agents?"}, {"id": 786594899, "item_id": 49162181, "url": "https://news.ycombinator.com/item?id=49162181", "by": "dregitsky", "date": "2026-08-03", "stance": "recommend", "quote": "Firecracker VMs + support for \"forking\" the box were our key requirements.", "excerpt": "For boxes.dev we use E2B - they had everything we needed at the time. I'd love a standardized API / marketplace, but more just to try out new providers with differentiated features or pricing - we probably wouldn't be switching often / using multiple (unless customers asked us for more options). There are so many sandbox providers popping up! Our use case is pretty specific though - one persistent machine that you se", "context": "builds boxes.dev", "use_case": "VM isolation for agent boxes", "switched_from": "", "story": "Devtools must be open source"}]}], "caveats": [{"label": "Complex operations", "summary": "Commenters say the microVM route is more complicated and harder to operate than containers.", "quotes": [{"id": 743366976, "item_id": 46460436, "url": "https://news.ycombinator.com/item?id=46460436", "by": "ammmir", "date": "2026-01-01", "stance": "neutral", "quote": "first with Firecracker microVM", "excerpt": "I've built the same thing twice, first with Firecracker microVM, and second time using containers (gVisor). While the microVM route is more secure, it's more complicated and ops are tricky, but you can do some cool things to optimize startup time like when I was working on a function as a service platform, and to reduce TTFB, I trapped the `listen()` call, sent a VSOCK message to the VMM to trigger a freeze, snapshot", "context": "built sandbox platforms", "use_case": "sandboxing / function as a service", "switched_from": "", "story": "Ask HN: What is the best microVMs for AI agents?"}, {"id": 762627267, "item_id": 47664204, "url": "https://news.ycombinator.com/item?id=47664204", "by": "benswerd", "date": "2026-04-06", "stance": "recommend", "quote": "For hardware virtualized machines it much harder but you can do it via:", "excerpt": "Self hosting can be doable for constant small/medium size workloads You can handroll a lot with: https://github.com/nestybox/sysbox?tab=readme-ov-file https://gvisor.dev https://github.com/containers/bubblewrap?tab=readme-ov-file For hardware virtualized machines it much harder but you can do it via: https://github.com/firecracker-microvm/firecracker/ https://github.com/cloud-hypervisor/cloud-hypervisor Freestyle/oth", "context": "", "use_case": "hardware-virtualized sandboxes", "switched_from": "", "story": "Launch HN: Freestyle: Sandboxes for AI Coding Agents"}]}, {"label": "Kubernetes integration", "summary": "One user found configuring it with k3s painful.", "quotes": [{"id": 789373106, "item_id": 49335819, "url": "https://news.ycombinator.com/item?id=49335819", "by": "srcreigh", "date": "2026-08-17", "stance": "recommend", "quote": "It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "excerpt": "Yeah. MicroVMs make hosting runners pretty straightforward. I had codex make a forgejo runner controller which pulls forgejo actions and creates kubernetes jobs for them. It was a bit of a pain to configure firecracker with k3s. It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "context": "", "use_case": "microVM CI runners", "switched_from": "", "story": "Ask HN: Alternatives to GitHub"}]}, {"label": "Linux only", "summary": "It is described as a Linux-only choice, with other tools suggested for cross-platform needs.", "quotes": [{"id": 789538194, "item_id": 49346137, "url": "https://news.ycombinator.com/item?id=49346137", "by": "Bnjoroge", "date": "2026-08-18", "stance": "recommend", "quote": "the typical choices are firecracker-based solutions", "excerpt": "If you want something cross-platform, smolvm or microsandbox are great options(built on top of libkrun vmm)! If you are looking for linux-only, the typical choices are firecracker-based solutions.", "context": "", "use_case": "Linux-only microVMs", "switched_from": "", "story": "Ask HN: Alternatives to GitHub"}]}, {"label": "Bare-bones", "summary": "One commenter describes another project as a more batteries-included alternative to it.", "quotes": [{"id": 787852320, "item_id": 49240770, "url": "https://news.ycombinator.com/item?id=49240770", "by": "binsquare", "date": "2026-08-10", "stance": "compare", "quote": "batteries included alternative to firecracker", "excerpt": "It's a batteries included alternative to firecracker with a couple of new ideas tossed into the mix i.e. portable like a container (bake into a single file and rehydrate the vm anywhere), dynamic resource allocation, etc.", "context": "", "use_case": "microVM sandboxes", "switched_from": "", "story": "Docker Sandboxes \u2013 Disposable, isolated sandboxes for AI agents"}]}], "who": [{"by": "dregitsky", "context": "builds boxes.dev", "url": "https://news.ycombinator.com/item?id=49162181", "date": "2026-08-03"}, {"by": "Reformedot", "context": "works on a browser infrastructure provider", "url": "https://news.ycombinator.com/item?id=48576698", "date": "2026-06-17"}, {"by": "_pdp_", "context": "runs CI runners on a single large server", "url": "https://news.ycombinator.com/item?id=48481684", "date": "2026-06-10"}, {"by": "spwa4", "context": "ML infrastructure", "url": "https://news.ycombinator.com/item?id=48291220", "date": "2026-05-27"}, {"by": "adamcharnock", "context": "runs a cloud provider alternative company", "url": "https://news.ycombinator.com/item?id=46948756", "date": "2026-02-09"}, {"by": "ammmir", "context": "built sandbox platforms", "url": "https://news.ycombinator.com/item?id=46460436", "date": "2026-01-01"}], "tags": ["for CI runners", "for AI agent sandboxes", "self-hostable", "for Linux users"], "use_cases": ["agent sandboxing", "microvm isolation", "vms for mobile agent platforms", "linux-only microvms", "microvm ci runners", "isolation"], "compared": [{"slug": "gvisor", "name": "gVisor", "co": 7, "switched_from": 0, "switched_to": 1}, {"slug": "docker", "name": "Docker", "co": 4, "switched_from": 2, "switched_to": 0}, {"slug": "unikraft", "name": "Unikraft", "co": 2, "switched_from": 1, "switched_to": 0}, {"slug": "kata-containers", "name": "Kata Containers", "co": 4, "switched_from": 0, "switched_to": 0}, {"slug": "podman", "name": "Podman", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "smolvm", "name": "smolvm", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "e2b", "name": "E2B", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=46450931", "title": "Ask HN: What is the best microVMs for AI agents?", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=49331033", "title": "Ask HN: Alternatives to GitHub", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=48468370", "title": "Surprise Pay $1000", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49965118", "title": "Linux containers in 500 lines of code (2016)", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49605644", "title": "The VMs Powering Mobile Agents (Instinct, Claude Code)", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49289512", "title": "Choose Boring Technology (2015)", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {"neutral": 2, "recommend": 1}, "2026-02": {"recommend": 2}, "2026-03": {}, "2026-04": {"recommend": 1, "neutral": 1}, "2026-05": {"compare": 1, "recommend": 1}, "2026-06": {"neutral": 1, "recommend": 1}, "2026-07": {"compare": 1, "recommend": 1}, "2026-08": {"recommend": 3, "compare": 2}, "2026-09": {"neutral": 1}, "2026-10": {"recommend": 1}}, "quotes": [{"id": 799520833, "item_id": 49970052, "url": "https://news.ycombinator.com/item?id=49970052", "by": "johnsmith1840", "date": "2026-10-05", "stance": "recommend", "quote": "Firecracker and gvisor are nice systems not horrible to use", "excerpt": "I've deploy gvisor, done basic test of firecracker and an honest attempt at production kata. Firecracker and gvisor are nice systems not horrible to use, gvisor isn't quite the same security level though. Kata is HARD to make. The technical know how to make that in production is awe inspiring. I wanted to use it but it was so complicated to integrate into a cluster I literally just gave up and mirrored raw VMs into t", "context": "", "use_case": "microVM isolation", "switched_from": "", "story": "Linux containers in 500 lines of code (2016)"}, {"id": 793690320, "item_id": 49605645, "url": "https://news.ycombinator.com/item?id=49605645", "by": "RohanAdwankar", "date": "2026-09-08", "stance": "neutral", "quote": "they use firecracker with memory in git", "excerpt": "I took a look inside the VMs that a couple of the agent platforms are using. TLDR is they use firecracker with memory in git. I'm a big fan of the memory in git part it would be cool to see how the git history looks after I use it for a longer period of time. This is my first blog, I'm trying to document things publically as I learn, next I'll look more closely at the providers they are using under the hood (eg. e2b)", "context": "", "use_case": "VMs for mobile agent platforms", "switched_from": "", "story": "The VMs Powering Mobile Agents (Instinct, Claude Code)"}, {"id": 789538194, "item_id": 49346137, "url": "https://news.ycombinator.com/item?id=49346137", "by": "Bnjoroge", "date": "2026-08-18", "stance": "recommend", "quote": "the typical choices are firecracker-based solutions", "excerpt": "If you want something cross-platform, smolvm or microsandbox are great options(built on top of libkrun vmm)! If you are looking for linux-only, the typical choices are firecracker-based solutions.", "context": "", "use_case": "Linux-only microVMs", "switched_from": "", "story": "Ask HN: Alternatives to GitHub"}, {"id": 789373106, "item_id": 49335819, "url": "https://news.ycombinator.com/item?id=49335819", "by": "srcreigh", "date": "2026-08-17", "stance": "recommend", "quote": "It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "excerpt": "Yeah. MicroVMs make hosting runners pretty straightforward. I had codex make a forgejo runner controller which pulls forgejo actions and creates kubernetes jobs for them. It was a bit of a pain to configure firecracker with k3s. It really can\u2019t be understated how much easier hosting CI is with microvms as the security boundary.", "context": "", "use_case": "microVM CI runners", "switched_from": "", "story": "Ask HN: Alternatives to GitHub"}, {"id": 788790818, "item_id": 49299426, "url": "https://news.ycombinator.com/item?id=49299426", "by": "insanitybit", "date": "2026-08-14", "stance": "compare", "quote": "the only two options we considered viable were Firecracker and gVisor", "excerpt": "I don't have a technology stack. Do you just mean what am I using for current projects? At work we use Rust primarily for services. It was chosen because we're doing a lot of low level and performance sensitive work. The risk we discussed most was devs not knowing the language, which we decided to hedge against in various ways and accepted that risk. We use Postgres for a lot of data. We've used it a ton at the compa", "context": "", "use_case": "isolation", "switched_from": "", "story": "Choose Boring Technology (2015)"}, {"id": 787852320, "item_id": 49240770, "url": "https://news.ycombinator.com/item?id=49240770", "by": "binsquare", "date": "2026-08-10", "stance": "compare", "quote": "batteries included alternative to firecracker", "excerpt": "It's a batteries included alternative to firecracker with a couple of new ideas tossed into the mix i.e. portable like a container (bake into a single file and rehydrate the vm anywhere), dynamic resource allocation, etc.", "context": "", "use_case": "microVM sandboxes", "switched_from": "", "story": "Docker Sandboxes \u2013 Disposable, isolated sandboxes for AI agents"}, {"id": 786594899, "item_id": 49162181, "url": "https://news.ycombinator.com/item?id=49162181", "by": "dregitsky", "date": "2026-08-03", "stance": "recommend", "quote": "Firecracker VMs + support for \"forking\" the box were our key requirements.", "excerpt": "For boxes.dev we use E2B - they had everything we needed at the time. I'd love a standardized API / marketplace, but more just to try out new providers with differentiated features or pricing - we probably wouldn't be switching often / using multiple (unless customers asked us for more options). There are so many sandbox providers popping up! Our use case is pretty specific though - one persistent machine that you se", "context": "builds boxes.dev", "use_case": "VM isolation for agent boxes", "switched_from": "", "story": "Devtools must be open source"}, {"id": 782301605, "item_id": 48893850, "url": "https://news.ycombinator.com/item?id=48893850", "by": "westurner", "date": "2026-07-13", "stance": "recommend", "quote": "that might be better for agent sessions: gVisor, firecracker vm", "excerpt": "(Years ago I had puppet and cobbler provisioning VMs over PXE and then iPXE. FWIU foreman is more actively maintained than cobbler, which is built on Django web framework.) Vagrant manages VMs and virtual networks, in Ruby. ansible-molecule creates, converges, and destroys VM(s) and containers, in order to test ansible playbooks and ansible roles in clean build roots. podman machine manages VMs: - podman-container-to", "context": "", "use_case": "agent session isolation", "switched_from": "", "story": "Show HN: Clawk \u2013 Give coding agents a disposable Linux VM, not your laptop"}, {"id": 780447696, "item_id": 48777981, "url": "https://news.ycombinator.com/item?id=48777981", "by": "3eb7988a1663", "date": "2026-07-03", "stance": "compare", "quote": "Do I have to go full, fat VMs or can I get by with a Firecracker-like thing?", "excerpt": "Related - what is the best isolation system available? Do I have to go full, fat VMs or can I get by with a Firecracker-like thing? Seemingly every available option has some subtle-gotchas about how easy it is to blow off your foot and effectively have no security at all. I use VMs because I actually trust that security is a foundational principle of the technology, not a well-if-you-use-these-20-flags-and-squint kin", "context": "", "use_case": "isolation for running agents", "switched_from": "", "story": "Jamesob's guide to running SOTA LLMs locally"}, {"id": 777227168, "item_id": 48576698, "url": "https://news.ycombinator.com/item?id=48576698", "by": "Reformedot", "date": "2026-06-17", "stance": "recommend", "quote": "we moved to a fully in-house solution with Firecracker and auto-scaling on EC2", "excerpt": "Our previous solution (Unikraft) did not supported auto-scaling That's why we moved to a fully in-house solution with Firecracker and auto-scaling on EC2", "context": "works on a browser infrastructure provider", "use_case": "browser VM infrastructure", "switched_from": "Unikraft", "story": "How We Run Firecracker VMs Inside EC2 and Start Browsers in <1s"}, {"id": 775706944, "item_id": 48481684, "url": "https://news.ycombinator.com/item?id=48481684", "by": "_pdp_", "date": "2026-06-10", "stance": "recommend", "quote": "we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "excerpt": "Haha, we don't use Kubernetes. It is a single large server with plenty of space for runners. It is as simple and boring as it gets. The only fun part is that we use Firecracker for the actual runners. It is provides nice isolation and it is very easy to work with.", "context": "runs CI runners on a single large server", "use_case": "isolated CI runners", "switched_from": "", "story": "Surprise Pay $1000"}, {"id": 775616562, "item_id": 48476035, "url": "https://news.ycombinator.com/item?id=48476035", "by": "_pdp_", "date": "2026-06-10", "stance": "neutral", "quote": "a Go service that orchestrates firecracker vms", "excerpt": "I am sure Blacksmith is the easy path if you don't want to do any extra work and I am also sure their solution will scale much better as this is what they do full-time. However, I can share what we did to ease on our GitHub Actions bills if it helps. Effectively we have our own runners hooked so that a job is scheduled, a runner picks it up and goes with it. We still use GitHub Actions but our monthly bill is now fla", "context": "", "use_case": "pre-warmed CI runner VMs", "switched_from": "", "story": "Surprise Pay $1000"}, {"id": 772659520, "item_id": 48291220, "url": "https://news.ycombinator.com/item?id=48291220", "by": "spwa4", "date": "2026-05-27", "stance": "recommend", "quote": "I should have switched to running docker containers in Firecracker looooong ago.", "excerpt": "The thing I find infuriating about docker is once you try to use it for serious machine learning, you absolutely require mounting your data. On a small scale we're talking s3 mounts, for large scale it's ceph mounts or (please don't) NFS. Oh and it must be an actual mount because speed. Or to put it differently: if you don't use kernel mounts your AI engineers need to design their training loop 10x better. It can wor", "context": "ML infrastructure", "use_case": "running ML containers", "switched_from": "Docker", "story": "Building an AWS Lambda-Like Runtime with Firecracker MicroVMs"}, {"id": 771143393, "item_id": 48196462, "url": "https://news.ycombinator.com/item?id=48196462", "by": "yjftsjthsd-h", "date": "2026-05-19", "stance": "compare", "quote": "Has anyone had luck moving more things to run under gvisor or firecracker or such?", "excerpt": "> We recommend exploring ways to automate upgrading your kernel Like, running emerge -u @world on a regular basis, or ... /me searches Okay, so https://wiki.gentoo.org/wiki/Live_patching exists but says, > A note of caution: Kernel live patching is risky. Count on hard freezing or panics to become normal... That's not encouraging. --- Another approach: Can we make the kernel vulns less important? Has anyone had luck ", "context": "Gentoo user", "use_case": "mitigating kernel vulnerabilities", "switched_from": "", "story": "Gentoo News: Copy Fail, Dirty Frag, and Fragnesia Kernel Vulnerabilities"}, {"id": 764947521, "item_id": 47809220, "url": "https://news.ycombinator.com/item?id=47809220", "by": "sahil-shubham", "date": "2026-04-17", "stance": "neutral", "quote": "on top of firecracker", "excerpt": "Nice work on Shuru \u2014 I remember looking at it when I was researching this space. You went with a Rust wrapper on Apple\u2019s Virtualization framework right? I have been working on something similar but on top of firecracker, called it bhatti (https://github.com/sahil-shubham/bhatti). I believe anyone with a spare linux box should be able to carve it into isolated programmable machines, without having to worry about provi", "context": "", "use_case": "microVM base for bhatti", "switched_from": "", "story": "Show HN: Smol machines \u2013 subsecond coldstart, portable virtual machines"}, {"id": 762627267, "item_id": 47664204, "url": "https://news.ycombinator.com/item?id=47664204", "by": "benswerd", "date": "2026-04-06", "stance": "recommend", "quote": "For hardware virtualized machines it much harder but you can do it via:", "excerpt": "Self hosting can be doable for constant small/medium size workloads You can handroll a lot with: https://github.com/nestybox/sysbox?tab=readme-ov-file https://gvisor.dev https://github.com/containers/bubblewrap?tab=readme-ov-file For hardware virtualized machines it much harder but you can do it via: https://github.com/firecracker-microvm/firecracker/ https://github.com/cloud-hypervisor/cloud-hypervisor Freestyle/oth", "context": "", "use_case": "hardware-virtualized sandboxes", "switched_from": "", "story": "Launch HN: Freestyle: Sandboxes for AI Coding Agents"}, {"id": 755121698, "item_id": 47195106, "url": "https://news.ycombinator.com/item?id=47195106", "by": "buremba", "date": "2026-02-28", "stance": "recommend", "quote": "use proper VMs like Kata Containers and Firecrackers.", "excerpt": "My take is that agents should only take actions that you can recover from by default. You can gradually give it more permission and build guardrails such as extra LLM auditing, time boxed whitelisted domains etc. That's what I'm experimenting with https://github.com/lobu-ai/lobu 1. Don't let it send emails from your personal account, only let it draft email and share the link with you. 2. Use incremental snapshots an", "context": "", "use_case": "isolating agents in production", "switched_from": "", "story": "Don't trust AI agents"}, {"id": 754508051, "item_id": 47156753, "url": "https://news.ycombinator.com/item?id=47156753", "by": "buremba", "date": "2026-02-25", "stance": "recommend", "quote": "use proper VMs like Kata Containers and Firecrackers.", "excerpt": "Sandboxes are not enough but you can have more observability into what the agent is doing, only give it access to read-only data and let it take irreversible actions that you can recover from. Here are some tips from building sandboxed multi-tenant version of Openclaw, my startup: https://github.com/lobu-ai/lobu 1. Don't let it send emails from your personal account, only let it draft email and share the link with yo", "context": "", "use_case": "agent sandboxing", "switched_from": "", "story": "Sandboxes won't save you from OpenClaw"}, {"id": 752686593, "item_id": 47042912, "url": "https://news.ycombinator.com/item?id=47042912", "by": "buremba", "date": "2026-02-16", "stance": "recommend", "quote": "I use Kata Containers on Kubernetes (Firecrackers)", "excerpt": "Neat! I wasn\u2019t aware that Docker has an embedded microVM option. I use Kata Containers on Kubernetes (Firecrackers) and restrict network access with a proxy that supports you to block/allow domain access. Also swap secrets at runtime so agents don\u2019t see any secrets (similar to Deno sandboxes) If anybody is interested in running agents ok K8S, here is my shameless plug: https://github.com/lobu-ai/lobu", "context": "", "use_case": "agent sandboxing", "switched_from": "", "story": "Running NanoClaw in a Docker Shell Sandbox"}, {"id": 751180097, "item_id": 46948756, "url": "https://news.ycombinator.com/item?id=46948756", "by": "adamcharnock", "date": "2026-02-09", "stance": "recommend", "quote": "We've created a Firecracker-based runner, which will run CI jobs in Firecracker VMs.", "excerpt": "We've migrated to Forgejo over the last couple of weeks. We position ourselves[0] as an alternative to the big cloud providers, so it seemed very silly that a critical piece of our own infrastructure could be taken out by a GitHub or Azure outage. It has been a pretty smooth process. Although we have done a couple of pieces of custom development: 1) We've created a Firecracker-based runner, which will run CI jobs in ", "context": "runs a cloud provider alternative company", "use_case": "CI job VMs", "switched_from": "", "story": "It's not you; GitHub is down again"}, {"id": 743380561, "item_id": 46461285, "url": "https://news.ycombinator.com/item?id=46461285", "by": "h1r1ms", "date": "2026-01-01", "stance": "neutral", "quote": "E2b also uses firecracker.", "excerpt": "Its good, also perplexity, gork some more startups uses E2B. https://e2b.dev/ have you tried this one? E2b also uses firecracker.", "context": "", "use_case": "microVM backend", "switched_from": "", "story": "Ask HN: What is the best microVMs for AI agents?"}, {"id": 743370032, "item_id": 46460627, "url": "https://news.ycombinator.com/item?id=46460627", "by": "zfoong", "date": "2026-01-01", "stance": "recommend", "quote": "We were thinking about using Docker and eventually settled on Firecracker.", "excerpt": "Thank you for your guidance! We were thinking about using Docker and eventually settled on Firecracker. Also, an interesting project you got there. If you are interested, would it be possible to invite you over to our project Discord? Would love to hear more of your experience.", "context": "", "use_case": "microVMs for AI agents", "switched_from": "Docker", "story": "Ask HN: What is the best microVMs for AI agents?"}, {"id": 743366976, "item_id": 46460436, "url": "https://news.ycombinator.com/item?id=46460436", "by": "ammmir", "date": "2026-01-01", "stance": "neutral", "quote": "first with Firecracker microVM", "excerpt": "I've built the same thing twice, first with Firecracker microVM, and second time using containers (gVisor). While the microVM route is more secure, it's more complicated and ops are tricky, but you can do some cool things to optimize startup time like when I was working on a function as a service platform, and to reduce TTFB, I trapped the `listen()` call, sent a VSOCK message to the VMM to trigger a freeze, snapshot", "context": "built sandbox platforms", "use_case": "sandboxing / function as a service", "switched_from": "", "story": "Ask HN: What is the best microVMs for AI agents?"}]}