zackify

5 ms·

1. 31.▲
   by zackify 3mo ago

   Same. Btop is the best
2. 32.▲
   by zackify 3mo ago

   I got mine! Did the reserve list 10 minutes in a few weeks ago. I like it way more than I thought.
3. 33.▲
   by zackify 3mo ago

   This is an annoying problem. It keeps making fake assumptions just because of hypothetical questions I've asked in the past. It'll assume I own a datacenter and have lots of gpus just because I asked to research things.
4. 34.▲
   by zackify 3mo ago

   You can get amazing local STT using parakeet which can use as little as 600mb of vram. Better or as good as whisper v3 large
5. 35.▲
   by zackify 3mo ago

   I got around 17m tokens on glm 5.2 then blocked for 4 days on the weekly limit on that plan.
6. 36.▲
   by zackify 3mo ago

   I did this by having a 30ft Bluetooth dongle in my attic near my living room
7. 37.▲
   by zackify 3mo ago

   I have moonlight and sunshine and also drilled through the other side of the house with a long fiber optic HDMI cable. Best of everything lol Also a Bluetooth dongle in my attic about 40ft USB cable. Works great for home assistant too and B
8. 38.▲
   by zackify 4mo ago

   I ran glm 5.2 on rented 8x h200 it could only do 2x concurrency at a cost of $40 an hour. It felt great but dang I wish it was cheaper... It needs 750 at fp8
9. 39.▲
   by zackify 4mo ago

   Calendly? Also google has this built in too now
10. 40.▲
    by zackify 4mo ago

    Yeah this is why I disable remote access and setup tailscale. Its annoying but with Claude and a little knowledge you can make it persistent. By default it got wiped every update which was annoying.
11. 41.▲
    by zackify 4mo ago

    I just approach everything as a one off task. Fresh context. "Use this CLI tool and figure it out. Look up this sentry issue using it" "Add a service that watches for an error in the log. Look at the home assistant MCP data a
12. 42.▲
    by zackify 4mo ago

    If only they would support the web and let you just issue a long running cookie.... I hacked the spec to pass through a cookie via the oauth handshake to do this without needing an oauth server. Its really dumb they don't want to allow
13. 43.▲
    by zackify 4mo ago

    pi.dev and ask ai to add features you miss from claude or codex. i configure keyboard shortcuts and swap models easily
14. 44.▲
    Ask HN: Why aren't hardware passkeys used for access token creation?

    4 points by zackify 4mo ago|0 comments
15. 45.▲
    by zackify 4mo ago

    You have to try pi.dev you can already make it do anything you want. I use opus to customize and tweak parts of it. Its the best harness due to the entire thing being api driven for customization
16. 46.▲
    by zackify 4mo ago

    Microcenter is the easiest place but almost any vendor will sell to you after you email them and if you have an LLC
17. 47.▲
    by zackify 4mo ago

    That's literally every project around AI. All the agent sandboxes. Hosting cron jobs that just hit ai rest endpoints for model completions etc
18. 48.▲
    by zackify 4mo ago

    Claude code plan mode. But yeah
19. 49.▲
    by zackify 4mo ago

    I have to share this because I thought it is behind funny how bad fable is doing at a task I JUST had opus do a week ago. it's also not even complicated: Copy my ssd to an external ssd so i can boot from it. Opus did this just fine. Fa
20. 50.▲
    by zackify 4mo ago

    Self made TUI that just lists LXC containers. I have a base container. "A" to make a new instance. Pi.dev when I hit enter on any container. Hot swap anthropic enterprise and openai and openrouter as needed. Every container has th
21. 51.▲
    by zackify 4mo ago

    Haha that's what I do personally. Vibe coded in 30mins a textualize tui that shows lxd containers. I just hit "p" on a container to forward that container to host. I only use ports for one instance at a time so it works perfe
22. 52.▲
    by zackify 4mo ago

    i can view the diff locally but often times after planning with opus i get what i want. I create a draft pr and manually review all items before then marking ready for review for the team. So I'm not blindly pushing things to prod with
23. 53.▲
    by zackify 4mo ago

    I vibe coded a TUI that just shows running lxd containers I hit 'n' to toggle all network access minus anthropic and openai URLs. I use pi (sometimes claude, always on bypass) and I auto allow everything. I only toggle manual appr
24. 54.▲
    by zackify 4mo ago

    As someone who negotiated 4 day weeks since early 2020 its been awesome. I get chores and yard work done and more family time every week. Wish it was standard.
25. 55.▲
    by zackify 4mo ago

    We are on it at my job. It saves money due to other parts of the org not using as many tokens. The real cost effective way is giving a team $20 cursor $20-100 Claude $20-200 codex. I'm spending 1k on Claude enterprise easily and that&#
26. 56.▲
    by zackify 5mo ago

    https://github.com/antirez/ds4
27. 57.▲
    by zackify 5mo ago

    Best deal currently: Cursor team Codex team Claude team Swap between the models when limited. I am saving our company a lot of money vs Claude enterprise usage cost
28. 58.▲
    by zackify 5mo ago

    Can they please do this with at&t internet.
29. 59.▲
    by zackify 5mo ago

    this thing is so awesome on fast mode, so far i am impressed, some of its observations feel similar to opus. i use gpt 5.5 and opus 4.7 a lot every day, if i can get good results at this speed, hopefully the usage level holds up on my team
30. 60.▲
    by zackify 5mo ago

    and this is why i use pi.dev and hotswap models and have no reliance on a single provider