{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiayudsobm3xvc7b6gziat3h4aipw7pqbx3fesdaj5sasflefizkaq",
    "uri": "at://did:plc:haakkg7y3xdghcdmprxeexso/app.bsky.feed.post/3mji5tbutpcd2"
  },
  "path": "/t/i-want-to-build-an-ai-llm-app-that-preserves-privacy-please-roast-my-idea/36714?page=2#post_26",
  "publishedAt": "2026-04-14T19:32:51.000Z",
  "site": "https://discuss.privacyguides.net",
  "textContent": "It’ll highly depend on your machine especially size of RAM. Since Qubes OS runs a lot of qubes (aka VMs) simultaneously there is some overhead for sure.\n\nI recommend at least 16 GB of RAM, the more the better but this depends on your use cases.",
  "title": "I want to build an AI/LLM app that preserves privacy. Please roast my idea :)"
}