BlockedPath

@1337hero try this flag on linux or mac export HF_HUB_DISABLE_XET=1

Why is downloading form Hugging Face so painfully slow?

@NousResearch Is this only for supergrok heavy?

SuperGrok now in Hermes Agent

Dont think ive had deepseek run in codex's harness yet

@loktar00 Location?

Hiring for an Associate Full Stack role, and we all know this market is nuts. Candidates are using AI to apply. I’m using AI to help review. It’s basically AI vs AI at this point. If your AI wins.... you win an interview 😂

@SpenceNicholson @unusual_whales End of may right? Time for me to get another gpu.

@BlockedPaths @unusual_whales subsidized tokens will be ending soon

"AI can cost more than human workers now," per Axios

YOU GUYS WOULD BE SURPRISED AT HOW MANY PEOPLE JUST SIT IN THEIR HOUSE. THAT’S IT. THAT’S THEIR LIFE.

That was a bar

@mil000 You guys aren’t ready for their operating system haha

Did they just kill 500 SSH apps that called themselves “codex on your phone”

@ClementDelangue @huggingface or just use the hard drive you already have. my models live on a 00 ubuntu box with a 0 SSD and 6gb vram. storage is free when you own the hardware

AI teams shouldn’t have to choose between expensive object storage and painful git workflows. @huggingface Storage is built for model weights, datasets, checkpoints and artifacts: - simple per-TB pricing - built-in CDN - Xet deduplication - private by default when needed Store your AI data where your AI work already happens: huggingface.co/storage

@HowToAI_ running locally is the whole point. i use qwen3 on my macbook for daily agent work. zero api costs, zero latency, zero surprise pricing changes. the repo is cool but local first should be the default not a feature

This repo shrinks 201GB of text down to 6GB without losing any accuracy. → 97% smaller than vector DBs → Runs locally → Drop-in MCP for Claude 100% open source. 100% private.

@mattpocockuk we dont need more acronyms we need agents that actually work without me having to write a whole readme just for them

TIL: DX: Developer Experience AX: Agent Experience AX is an awesome descriptor for something I've been thinking about - how well an agent can perform in your codebase How well-architected it is. How good the feedback loops are. How discoverable information is. Love it.

@OpenRouter or just use deepseek directly. /bin/sh.14 per million tokens and the quality is the same as the 5/million stuff. why mess with routing keys when the cheapest model already wins

Three big upgrades to BYOK on OpenRouter 🔑 1/ Add multiple keys for the same provider in one workspace and set the order they're tried in. Useful for rotating across rate limits, separating dev and prod credentials, or distributing usage across team accounts!

@Everlier @loktar00 Slowly become naive, comfortable.. until its too late lol

@BlockedPaths @loktar00 I think the trend recently is that if there's too much of it but none at all at the same time - that's fake

Building a tool to help catch stale skills

@Everlier @loktar00 Half the time you cant tell whats real online anymore, Im constantly thinking something is fake. hahaha

@BlockedPaths @loktar00 The whole ecosystem is too much for years already, I have a feeling some of my childhood memories are being replaced with knowledge of some of the AI projects

@SIGKITTEN @badlogicgames You can remove it or append it per projects in .pi

@badlogicgames its almost perfect! can u remove the system prompt too

sneak peak of the new pi.dev no tools mode. you're gonna love it. you can try it with `pi -nbt`. enjoy!

@1337hero 100% Its so simple at first and allows you take a slow pace and absord everything. Works very easy with local models over tailscale as well.

"Pi agent is the Arch Linux of coding agents" stealing this quote - banger

@nicos_ai Do any of these certs I see make a shit of difference?

GITHUB ACABA DE LANZAR LA CERTIFICACIÓN OFICIAL DE UNO DE LOS ROLES TECH MÁS IMPORTANTES DE 2026 → Agentic AI Developer (GH-600) Y es la primera vez que trabajar con agentes de IA se convierte oficialmente en una disciplina reconocida de ingeniería. Ya no hablamos de: • prompt engineering • vibe coding • automatizaciones simples Hablamos de un nuevo perfil técnico: → Agentic AI Developer La persona que: • coordina agentes de IA • construye workflows autónomos • integra agentes en entornos reales • supervisa fallos en producción • evita errores críticos en pipelines CI/CD • sabe cuándo un agente no es fiable Antes: → “Trabajo con agentes de IA” era difícil de validar. Ahora: → GitHub certifica oficialmente ese skillset. Y eso cambia el mercado. Las empresas van a necesitar este perfil. Pero todavía hay muy pocos developers especializados en ello. Si ya trabajas con: • Copilot • Codex • Claude Code • workflows agentic • automatizaciones con IA Probablemente ya estés haciendo este trabajo. GH-600 es la forma de demostrarlo. Guárdate esto 🔖

@digitalix @dcapitella You’re the goat

I spent months trying to make this Strix Halo AI cluster work the way I thought it should. (new video linked below) @dcapitella had the key

@Gavel_on_X The grind is hard

Daily engagement goal on X: reached! ✅ Thanks, nerds! 😴

@loktar00 @Everlier Skills, extension, plugins, prompts, it getting to be too much 😩

@Everlier Missed opportunity to call it Skill Issue