Igor Alentyev

@samsheffer Yeah really x.com/JSFILMZ0412/st…

i don’t think you understand how insane omni is

@jrfetzer @the_smart_ape What bout diliverability?

@aindienomad @the_smart_ape Private Exchange Server

> use a google account for 15+ years > do everything with it: youtube: google, drive, playstore > launch a saas > stripe, slack, notion, vercel, ....: sign in with google > wake up and see: "your account has been suspended for ..." > can't open notion → 3 years of client files gone > can't open stripe → can't refund > can't reach cofounder (slack-sso-via-google) > domain renewal notice goes to dead inbox > lose a $140k mrr saas > still don't know what i did wrong > mfw "sign in with google is more secure than passwords"

@NotFifthGear bad looking golf

That's just a Golf

@lgs_pro @L0m3z 🤮

@L0m3z seen in the wild a few days afo

Shocking to me that companies are still using this art style

@lewiscarhart Or just Laravel

this is the ultimate tech stack nothing comes close 🤷‍♂️

@jrfetzer @the_smart_ape Name? g is easiest way to setup corpo email

@the_smart_ape Yep, this is why you use a true business class email service with a more business friendly SLA.

@rubenhume cringe

At some point you grow up and realize the only correct color for a MacBook is silver

@ashleyhindle Hey artisans, just install Reverb (with UV) or Soketi or whatever WS serever that tooks 20 minutes with claude for setup. That is really that simple and you will have nearly unlimited requests.

Laravel developers now get websockets setup in ~40 seconds, for free, using vask.dev and the new composer package 🔥 artisan vask:install will auth with vask.dev, give you a websocket app, and set those details in your .env file so you're ready to go 💪 Coming soon: your agent will register for you, no OAuth needed 😅

@thdxr For you guys - of course!

opencode go is currently zero data retention however we can increase limits and make it all more sustainable if we collect data to train future open source models you can opt out of this - is that something you'd be ok with?

@RhysSullivan I'm building in the HRtech domain and I accidentally found geekploy.com. It sounds nice, and a .com domain is rare.

Every good domain is taken

@ar0cket1 @art_zucker @IlysMoutawwakil also very useless for real-world tasks besides openclaw useless stuff

@art_zucker @IlysMoutawwakil MoEs are great, crazy token efficient and very cheap to run flop wise.

Today I re-iterate: I hate MoEs and we are wasting time on them.... Let's unite and call a global ban on MoEs please. Please 1M+ salary researchers: do better... credits to @IlysMoutawwakil for the graph:

@enunomaduro Never. Why? Mysql/Postgres installation is trivial

ho yeah baby.. by default

@LLMJunky @royjossfolk @SuperDesignDev also doubled for tool, using it for 2 months, quite cool.

@royjossfolk yeah bro. thank you 🫶 i used @SuperDesignDev underrated tool. doesn't get enough love. copy the prompt, tweak from there. this is the result of a lot of love.

Calling on Codex Fans! I need your help 🫵 I'm introducing Codex Marketplace, a community collection of plugins, skills, and hooks curated by YOU. Yes, you! Help me make this the best resource for everyone who wants to build incredible software with Codex. Getting started is simple, submit your artifacts via your Github repository. If you own the repo, it'll be auto approved, else they'll be reviewed. Upvote your favorite artifacts. Add/remove what you like with a simple npx command. Best part? Everything lives on Github, so you always get the latest version. 👉 codex-marketplace.com

@LLMJunky @royjossfolk @SuperDesignDev bro what prompt?

@jun_song so cope

2개월 뒤에는 당신의 맥북에서 Opus-4.6 수준의 로컬AI를 구동할수 있을겁니다. 그것의 이름은 Qwen3.7입니다. 이 트윗을 저장하고 정확히 2달뒤에 보세요.

@lucaronin It looks beautiful! I desperately need this for windows!

Introducing Tolaria! 💧 Today I am releasing a macOS desktop app for managing markdown knowledge bases, and helping both AI and humans operate them. It’s free and open source, and always will be. I have been working on it for three months, and I now use it to run my life and work. I personally have a massive workspace of 10,000 notes — the result of 6 years of Refactoring — which I now operate on Tolaria. Tolaria is the main collaboration surface with my AI agents: they create new notes there, connect them to what exists, and edit existing ones. Everything is easy to understand for them, because it’s just markdown files. In a way, it’s my implementation of @karpathy's LLM wiki. Tolaria is also the biggest experiment I have ever run about writing software with AI: • 2000 commits • 100K+ lines of code • 3000+ tests / 85% coverage • 9.9/10 code health • 70+ architecture decision records I am releasing it open source also to use it as a living artifact of how I do AI coding, so you can inspect at any time things like how I write docs, what's in my AGENTS file, what hooks do I run, and so on. You can find it below: • Newsletter announcement: refactoring.fm/p/introducing-… • Website: tolaria.md • Github repo: github.com/refactoringhq/… Let me know your thoughts!

@joshuamschultz Good job and thanks for the article! Hetzner is peak, so powerfull and cheap! Did the same thing for hirify.me couple weeks ago. Its not so big as you but steps was pretty close!

Someone just migrated from DigitalOcean to Hetzner. Same production workload. 30 MySQL databases. 248GB of data. Hundreds of thousands of users. Bill went from $1,432/month to $233/month. New machine is MORE powerful. 96 CPUs. 256GB DDR5. NVMe RAID. $14,388 saved per year. This is happening everywhere.

@nicdunz It is SkiBiDi architecture for real

absolutely groundbreaking

@niccruzpatane who the fuck placing new on the left side?

Volkswagen revealed the interior of its new ID.3 Neo yesterday. VW’s CEO said the previous model’s interior (right image) "definitely did a lot of damage" to the brand due to the lack of buttons and physical controls. In the new ID.3, VW has added back physical climate control buttons, a volume knob, more buttons on the steering wheel, and actual window switches.

@MrPunyapal Opus is x3 in requests, new opus is x7.5 Gpt5.4 is x1 nonsense.

Unpopular opinion: with GitHub Copilot you can do same work using same model (Opus, Sonnet) and consume less credits than Claude Code.