Live data from Hacker News

Viewing profile — fariszr

fariszr

HN member
Joined
Wed, Jul 13, 2022, 10:17 PM UTC
HN karma
1,665
Public activity
323 items

About fariszr

https://fariszr.com/about

Recent public activity

  1. comment
    Comment #49097200

    But does joblin support anything more than markdown? I'm thinking of using it as a like a store for agent plans, so I can easily edit stuff. Can an agent add rendered HTML, meramai…

  2. story
  3. comment
    Comment #46302027

    Sure, but for simple tasks that require a large context window, aka the typical usecase for 2.0 flash, it's still significantly more expensive.

  4. comment
    Comment #46301917

    These flash models keep getting more expensive with every release. Is there an OSS model that's better than 2.0 flash with similar pricing, speed and a 1m context window? Edit: thi…

  5. story
  6. comment
    Comment #45965656

    I mean over in that I don't see a need to use the other models. Codex models are the best but incredibly slow. Claude models are not as good(IMO) but much faster. If gemini can bea…

  7. comment
    Comment #45965530

    This is a big jump in most benchmarks.And if it can match other models in coding while having that Google TPM inference speed and the actually native 1m context window, it's going …

  8. comment
    Comment #45870868

    https://github.com/FarisZR/knocker Knocker, an http knock based access service for your homelab that works at a reverse proxy or firewall level. It's a more convenient albeit less …

  9. comment
    Comment #45753510

    Is there a way to force clients to use a relay? It seems like this is only meant as a fallback, but what if a relayed connection is actually faster (like when direct peering betwee…

  10. comment
    Comment #45676795

    Hey I'm the creator of knocker! I actually wanted to write a blog post about it before posting, but OP already did that. If you have any questions just let me know! Will go into mo…

  11. comment
    Comment #45676749

    The TTL is for the whitelist. The whitelist rules aren't permanent.

  12. comment
    Comment #45676746

    Sorry if i wasn't clear. It isn't more secure, it's just more convenient because it works in every network, without needing to set up a VPN connection on each device. I created thi…

  13. comment
    Comment #45676726

    > Why not just run a reverse proxy in front of (whatever service you're trying to protect) and use the API keys there? Because it breaks the clients of most homelab services. That'…

  14. comment
    Comment #45676716

    It's a compromise.It's not as secure as using a VPN, but it's way more convenient, since only one device has to have a knocker client on it without needing any sort of VPN. The lik…

  15. comment
    Comment #45676680

    Tailscale is not as easy as this. It has to be installed on every device or at the router level. And it will not work on mobile if you already use another VPN.

  16. comment
    Comment #45676663

    I mean you are free to not use it, it's for personal use. I was annoyed by all the vpn based solutions and built knocker to have something that works without installing it on each …

  17. comment
    Comment #45676658

    I mean it's just using firewalld. You can't inspect the rules. For me it's simple enough that it shouldn't be a big security issue, but I understand and that's why I wrote that in …

  18. comment
    Comment #45676649

    I use tailscale for this. But I can't have two vps running at the same time on android, nor do I want to install tailscale on every device I own. I created knocker exactly for this…

  19. comment
    Comment #45590108

    Finally! Pixel hardware is a joke, the pixel 10 pro has the performance of a three year old phone, with battery life worse than the iPhone Air (according to shortcircut/ltt tests).…

  20. comment
    Comment #45428260

    Did they make human voices sound robotic on purpose? Is that some kind of Ai fingerprinting? It's way too obvious

  21. story
  22. story
  23. comment
    Comment #45032306

    The phone is just in general sluggish, but video calls and usage on cellular are a catastrophe, it just turns into a hot potato.

  24. comment
    Comment #45032273

    Tool use and sycophancy are still big issues in gemini 2.5 models.

  25. comment
    Comment #45031918

    I'm sorry I absolutely don't agree. This model is on a whole other level. It's not even close. https://twitter.com/fareszr/status/1960436757822103721