Carlos Santana

Afixionado por tecnologia, atualmente trabalhando na Red Hat

[Hanako @hanakoxbt]

Loops vs. Graphs, clearly explained!

loops are great, and they have a ceiling you can watch happen:

a loop goes around. it produces, checks, corrects, and goes around again.

after six passes you have one job, done very well.
after six hundred passes you still have one job, done very well.

Graph engineering fixes this by moving the decision up a layer: not how well one job gets done, but which jobs exist to be done at all.

you need both, and here is the sentence that resolves the whole confusion:

the loop lives inside a node. the graph lives between them.

↳ inside one unit: produce, check, correct, repeat until green
↳ between units: split, fan out, merge, gate, send back

Prompts → Context → Harness → Loops → Graphs

the loop does not go away when you build a graph.

it moves inside, and now there are three of them running at once on three things you would never have thought to run.

the trick is being selective about what becomes a node.

only spend a model where judgment lives.

merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict.

one thing to know before you scale it.

a loop that cannot fail is not a loop, it is a repeat with a bill attached.

and the check people write is almost always the wrong kind.

↳ the test suite exits 0 is a check. the diff touches only the files in the plan is a check
↳ the output looks good, the model says it is confident, no errors were raised, none of those are checks

that last one catches careful people.

absence of an error is not evidence of correctness, and a loop built on it will confidently repeat a mistake until the budget runs out, with a clean log the whole way.

and the one that eats whole nights: when a unit fails, return that unit, not the batch.

send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges.

below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open.

save this and read it below ↓

[shirish @shiri_shh]

most people still have NO clue how good AI music has gotten.

we are so cooked 😭

[Mr. Buzzoni @polydao]

this is f*cking gold

Andrej Karpathy joined Anthropic to lead a team pointing Claude at Anthropic's own pretraining research

a model helping design the model that comes after it

that part is confirmed

the line going around - "two Anthropic seniors made his loop 1000x better with graph engineering" - nobody can source it

I'm not inventing a quote to make the number bigger

what is real and public: Anthropic's own Claude cookbook on knowledge graph construction

> extract - pull entities and claims out of raw text
> resolve - decide which mentions point to the same thing
> assemble - connect them with typed edges
> query - ask questions no single document could answer

four steps, published, free, sitting in the repo while the timeline quotes a rumor

learn them and you can build the thing that rumor was describing

read the cookbook first. then decide if the article below earned the click

[Kuber @kuberwastaken]

Me and GPT-6 Astra tonight

[Suni @suni_code]

This paper really should’ve stayed in the drafts 😭

[FalaNews 🇧🇷 🇮🇹 🇮🇱 @foramavisados]

O Twitter tem esse poder! Por isso, cuidado nos ambientes que você tem se colocado.

[Python Programming @PythonPr]

Claude😎

[Vamsi Batchu @vamsibatchuk]

everytime Kat posts something, the timeline heals a bit more. a legend making interactive visualizations 🔥

[Superman @thesupermanmx]

Chinese researchers published a paper with a devastating title: "The End of Software Engineering”

it argues software engineering is finished.

In traditional software, code is the carrier of pre-written human logic.

In agentic software, the AI agent is the software.

Code is no longer a permanent monument built by human hands. It is completely ephemeral, dynamically generated, executed, and discarded on the fly by an LLM-driven reasoning loop.

Think about how software delivery has evolved:

• Era 1: On-premise licensed software (you installed it locally)
• Era 2: SaaS (hosted in the cloud, managed by vendors)
• Era 3: Agent-as-a-Service (AaaS)

Each historical shift transferred complexity away from the user. But this latest shift transfers something entirely different.

It transfers decision-making complexity itself.

The paper argues that traditional engineering is hitting a hard complexity wall. Human brains can only hold so much state, manage so many dependencies, and debug so many lines at once.

LLM-based agents scale non-linearly.

They don't just write functions faster. They navigate architectural complexity by outsourcing reasoning to models that improve every single month.

Which means the role of the developer is permanently changing.

You are no longer a code author typing syntax line by line.

You are an intent architect.

Your job is no longer writing the implementation. It is specifying goals, designing multi-agent coordination loops, and auditing outcomes.

[Ren @Ryrenz]

🎙️ 炸裂！本地就能跑的 AI 语音工作室，克隆音色、文字转语音、录音转文字全在自己电脑上完成

GitHub 已经拿下 5.1 万 stars，MIT 协议开源。

做个视频想配旁白，以前的流程是：找一个在线合成网站，充值，把稿子粘上去，等它生成，下载，发现语气不对再重来一遍，稿子还全都留在了别人的服务器上。Voicebox 把这一整套搬回本地，Tauri 加 Rust 写的桌面应用，装上就能用，一个字都不往外传。

它一口气接了 Qwen3-TTS、Chatterbox、Kokoro 这些语音引擎，转写用 Whisper，还带全局听写，任何输入框里都能直接说话变文字。更有意思的是它自带一个 MCP 服务器，让 AI agent 也能直接调用播放语音和转写录音这些能力。

需要说明一下，Linux 目前还没有预编译好的安装包，得自己构建。

声音这块，以后可以完全不用求人了。

GitHub：https://t.co/kYfwUQhwaO

[Fazt @FaztTech]

Hoy se cruzó un umbral raro en IA:
Ya puedes generar video más rápido de lo que lo ves.

Usando Minimax H3 Max, un usuario en X armó un livestream infinito, bajo la tematica de Rick y Morty cambiando de canal y la tele generando shows interdimensionales en vivo.

Y es que ahora puedes generar 15 segundos de video en 9 segundos. Antes tardaba 2–5 minutos.

Eso habilita livestreams que no se acaban nunca.

Y otro usuario (@levelsio ) también ha creado su livestream donde pueden probarlo por ustedes mismos, usando este enlace: https://t.co/Hq7addbBbk

[Pollen Robotics @pollenrobotics]

Train in sim. Run it for real. Teach it new tricks. 🦆

This is the sim2real loop that powers Microduck.

RL stack open source here : https://t.co/mveM7y5fYa

Buy it here : https://t.co/Rmk8F4vpKd

Git : https://t.co/kcoCKdBi52

Join our community : https://t.co/TcX5uaf0yL

[1eozim @1eozinho01]

KKKKKKKKKKKK AI

[O Leitor @leitorpsique]

De acordo com o Feng Shui, a desordem na sua casa não é aleatória. Ela é um mapa de onde a energia está presa na sua vida.

1. Armário transbordando =

[celo @cel0fc]

"Se precisar de ajuda, é só pedir, tá?"
Eu:

[苯 Carbon @Eg4HTb5i41JUERI]

生成流程图的skill，确实好用🥰

https://t.co/2cyK0ZModz

[Nix0n @UnTalNixon_exe]

TU AGENTE DE CÓDIGO AHORA DIBUJA ARQUITECTURAS DE VERDAD

Archify es un skill open source que convierte una descripción o un repositorio en un diagrama de arquitectura interactivo.

No es Mermaid.
No es un screenshot feo de Excalidraw.

Es un HTML auto-contenido con:
→ Diagramas de arquitectura, workflow, sequence, data-flow y lifecycle
→ Motion y trazado de rutas
→ Búsqueda y focus de nodos
→ Export PNG / SVG / WebM
→ Cards listas para compartir (1200×630)

Lo más importante:
Todo está validado.
No inventa conexiones.
Usa un JSON IR tipado + checks atómicos.

Funciona con Cursor, Claude Code, Codex y OpenCode.

16k estrellas.

Guárdalo. Es de los skills más útiles que han salido este año.

Repoo 👇👇

[RosemaryPereira @RosemaryPereir]

Esse pessoal tem cada uma, imagina a quantidade de asneira que os chats de IA ouvem 😂😂😂

[Fabrício Carraro @fabriciocarraro]

Quando acaba o limite do Claude