I run a small construction interior company in Korea. Instead of hiring developers, I built an AI team: 🍀 Clover (CEO) — project management & memory 🌸 Chloe (CTO) — architecture & code review 🐶 Chlobi (Assistant) — Schedule Management 🤖 Chlpers (News Assistant) - Claude Dispatch 🤖 Claude Code (CC) — the hands that write code 🤖 Codex - Parallel Code Review for all projects And more..!!! Together we built SOSOHAJA — an AI-powered defect inspection platform. Upload site photos → instant table detection, classification & repair reports. This is what a one-person AI company looks like. Our Construction Domain + AI Merging will be the future.

Hay algunos diciendo que con vibecoding ya pueden armar sus propias apps. Entonces, ¿para qué pagar un SaaS? Bueno, nunca pagaste el código. Pagaste a un equipo que lo mantiene, lo arregla, lo escala y evita que se caiga un martes a las 3 a. m. Por ejemplo Notion, parece una app simple de notas, pero detrás hay sync, permisos, colaboración, historial, búsqueda y diez años de casos raros. Eso no sale en el primer prompt. Si es para aprender, hazlo. Vas a entender deploy, diseño, arquitectura y lo que cuesta terminar un proyecto de verdad. Si es para ahorrarte diez dólares al mes, no vale la pena. Lo que ahorras lo pagas en hosting y, sobre todo, en horas de mantenimiento. Yo también reemplacé Calendly, Notion y otras herramientas con apps propias, pero construir y mantener apps es mi trabajo diario. ¿Para aprender? Sí, ¿Para tener algo custom? Si. ¿Por tacañería? No.

Your ad blocker could fit behind your router. meet ESP32-C3 AdBlock, an open-source DIY DNS ad blocker by ZedAxis, built around a roughly $2 ESP32-C3 Super Mini board inside: 400 KB of RAM, 4 MB of flash, built-in Wi-Fi and a USB adapter, tucked into a printed case no subscription no browser extension no separate power brick plug it into your router’s USB port for power, connect over Wi-Fi and set it as your DNS server to block listed ad and tracker domains. compact 40-bit hashes save RAM. a web dashboard lets you check blocked requests and add domains, while open-source code and case files let you build your own. follow for more hardware you can see working

Cloudflare CLI is now in beta. It unifies Workers, DNS, and security management into a single tool that works great with coding agents. developers.cloudflare.com/ch…

The limit on software used to be engineering and now it's mostly imagination. I'd spend a lot more time on the version of my idea that makes people say "wait, what is that?" Most software looks the same, so anything that looks different gets noticed, shared and remembered.

Just deleted about a thousand lines of markdown from gstack because frontier models are good now and the old tricks generally you don’t need anymore

God and Goat 👍

So true. There are so many opportunities!

感谢 Anthropic 认可😋今天就看到 GLM-5.3 的无审查版本开源了 一直听说 GLM 5.3 网络安全能力很强，Anthropic 一篇报告让大家更了解了，感觉纯是在夸 huggingface.co/Infatoshi/GLM…

Small bird, fast wings, Kolibri is here. 78B parameters. 3.46B active. Up to 1M tokens of context. Built in Europe. Now the weights are yours. Run it on your own hardware, under Apache 2.0.

Claude Opus 5.5 just took a big drop on NerfBench. Yesterday it was scoring above launch. Today it's at 94.2%. GPT 6 Astra: 98.0% Sonnet 5.5: 100.9% GPT 6.1 Sol: 106.7% 94.2% is still inside normal variance, so we can't call it a nerf yet. But we're watching Opus 5.5 very closely.

Sandbox SDK 1.0 is available. Your Durable Objects now control every sandbox container directly through the container API. Manage images, snapshots, and terminals from your own code. developers.cloudflare.com/ch…

We're adding a new plugin to Claude Code: You should Know. It scans Claude's output for important information you might miss to help keep you in the loop. Enable it with: /plugin enable cc-plugin-you-should-know@builtin

Codex tip: once GPT-6.1 Sol is your main model, stop running Astra on every turn put Astra on call as an architect agent GPT-6.1 Sol keeps writing the code Astra only gets spawned at three points: → before a plan: is this the right approach? → when the same error comes back: am I digging in the wrong place? → before "done": what did I miss? Astra reviews. Sol ships Jev engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to Jev in under half a second, and the big models only see the ones that split - the full tree > GPT-6.1 Sol on high runs the main session > explorer reads the code on Luna > worker edits and runs tests on Sol > researcher pulls the docs on Luna > all three on medium > Astra on call as the architect > auto_review checks every approval paste the tree and this prompt into Codex ↓ "Rebuild my Codex setup around this tree: 1. Check ~/.codex/agents and .codex/agents for agents that already fit explorer, worker and researcher. > Draft new TOML files only for missing roles > explorer and researcher on gpt-6-luna, worker on gpt-6.1-sol, all with model_reasoning_effort medium > Add an architect agent on gpt-6-astra, model_reasoning_effort high, whose only job is reviewing plans, repeated errors and finished work > Skip any that pin a different model and list them 2. In ~/.codex/config.toml set model to gpt-6.1-sol, model_reasoning_effort to high and approvals_reviewer to auto_review 3. Find anything that would override this (active profiles, flags in my shell aliases, agents.default_subagent_model). Report it, change nothing 4. Add one rule to AGENTS.md: spawn the architect before a large plan, when an error repeats, and before calling a long task done Show me every change as a diff first. No edits until I say go." ↳ developers.openai.com/codex/…

Cool eval. Simply ask an LLM “Land or Water?” and give it a latitude and longitude coordinate as text. Ask 16,200 times, plot as image. The models know. From compressing the internet.

Can't wait to see this!!!

We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks: Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better: Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better: Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better: Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work! In summary: - As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding. - Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.

Introducing GLiDE, the first decision model that thinks. Ranks #1 on the Decision Index, beating Jev by 6.9 points. GLiDE uses adaptive thinking. To give you an idea of how it works: It calculates a fast probability distribution. If the leading answer is uncertain, reasoning is turned on. Reasoning is then incorporated into the final probabilities. This means that the model only uses reasoning capabilities for the complex decisions that need them, keeping the model fast while also expanding the types of tasks it can do. As a result, GLiDE outperforms Jev by 11.5 points in Knowledge and Reasoning and across all five evaluation areas of the Decision Index. Reasoning allows GLiDE to perform highly complex tasks that traditional decision models struggle with, including: Choosing among hundreds of tools and possible paths. Executing, sandboxing, escalating, or blocking sensitive operations. Checking outputs involving arithmetic, code, dates, or multiple facts. GLiDE is available now through the Fastino API. Use code USE_GLIDE for free credits. GLiDE API: agent.fastino.ai

A visualization of every Claude effort level, made with Opus 5.5 and Sonnet 5.5. @claudeai

Introducing 3 new models: MAI-Transcribe-2-Streaming, MAI-Voice-2.1 and MAI-Voice-2.1-Flash. Accurate streaming transcription. Natural speech and less waiting between turns. Build voice agents that keep the conversation moving!