zackify

$6 a month I plan to use deepseek v4 flash mainly which should provide closer to 5x the usage on the cheaper ones but no set number

I switched to yearly Cline pass because it was too cheap haha

Same. Btop is the best

I got mine! Did the reserve list 10 minutes in a few weeks ago. I like it way more than I thought.

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.

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

I got around 17m tokens on glm 5.2 then blocked for 4 days on the weekly limit on that plan.

I did this by having a 30ft Bluetooth dongle in my attic near my living room

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 Bluetooth devices like plant sensors outside the house over ble

Bonus points if you do tailscale and jetkvm or wake on LAN and can moonlight from anywhere.

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

Calendly? Also google has this built in too now

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.

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 and find my phone, send a notification to it and make it send when there's an error"

Now after it manually does it and makes the code. "Make this into a skill for anytime I paste a URL with sentry inside the message"

But also important: "make that procedure a prompt file so when I invoke it I just pass X after it and it works fully"

Having too many skills for things that are very specific where you can directly invoke it via slash command, wastes context space with the skill headers I find. So I make those prompts instead

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 this.

If no cookie, open webpage.

If cookie set, close and persist.

I literally wrote an 80 page mini book on MCP yet it frustrates me to no end.

pi.dev and ask ai to add features you miss from claude or codex. i configure keyboard shortcuts and swap models easily

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

Microcenter is the easiest place but almost any vendor will sell to you after you email them and if you have an LLC

That's literally every project around AI. All the agent sandboxes. Hosting cron jobs that just hit ai rest endpoints for model completions etc

Claude code plan mode. But yeah

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.

Fable planned to have me reboot to safe mode. ok thats fine. I told it no.

It started copying and overwriting the ssd while IN PLAN MODE. this is crazy it feels so dumb vs the marketing

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 the dev env already running for my current projects. Iterate, rarely use vim when needed, spec driven and have llm draft prs for me then I review.

I know the codebase in and out so what I want done is on bypass mode and then I review closer at the draft PR step before marking ready for the team.

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 perfect.

Hitting enter auto joins the lxc container instance with tmux.

Works perfect for me for tasks that can stay long running

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 without review.

Without staging key access I wouldn't have been able to do a payment provider migration at this speed. iterating by migrating users in staging and being able to use and validate the sdk quickly with opus is a massive time saver.

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 approval in rare cases like running a script or command that needs to touch a production system and I need to validate everything.

Normally my container has full write access to staging so it can debug and validate everything on its own

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.

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's with trying to spread it on codex and cursor using pi.

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

Can they please do this with at&t internet.

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 plan haha