OWΞZZY

While Alive: 💚 life ||💙 code: @Angular @reactjs @AWS @Ethereum || 🧡music &&
return owezzy;

[OWΞZZY @owen_adira]

@kasike Not yet. The agent did that on its own: put the old middleware back, ran the matrix, and then restored the fix. That was the strongest evidence in the handoff, so it’s going into the handoff format: a test added with a fix has to be shown failing without it.

[OWΞZZY @owen_adira]

I fixed what my agent left behind

Two months after a coding agent turned a failing test green and left an inconsistent API behind, I gave another agent the brief I should have written. It stopped three times, and each stop…

#ai #llm #architecture #system
https://t.co/b89ZA4yuTe

[OWΞZZY @owen_adira]

Where should a coding agent stop and ask?

Approval checkpoints for coding agents: which actions should pause for a human, what the approver needs to see, and how to match safeguards to risk without making low-risk work painful.

#ai #architecture #system
https://t.co/CKyglYnjOI

[OWΞZZY @owen_adira]

Show your work: the evidence a coding agent should hand back

An agent that returns a diff and a cheerful "Done!" moves the work to the reviewer. What evidence different tasks should return, which checkpoints keep long-running work…

#ai #llm #system
https://t.co/piV9dUBhNR

[OWΞZZY @owen_adira]

"Fix the build" is not a task: how to hand work to a coding agent

Autocomplete, pair programming and delegation need different amounts of structure. A five-point readiness check and a short delegation brief turn a vague request into…

#ai #llm #webdev
https://t.co/AEvZxUopmF

[Armin Ronacher ⇌ @mitsuhiko]

Opus Induced AI Psychosis https://t.co/GMn9i4c1Kb

[chasm @chasmmmmmmmmmmm]

NOT EVEN THE PLAYSTATION 5 IS SAFE 🚨

Someone jailbroke a PS5 and got Qwen 3.8 27B running on its GPU at 20 tok/s. That's a real 27B AI model running locally on a games console.

The same model on the PS5's CPU did about one token a minute.

Meanwhile, DDR4 prices are up 24% in 3 months, AMD just raised GPU memory prices, and 5090s are basically gone.

PS5 coming to a datacenter near you.

[SAMMY WASILWA @samuelwasilwa5]

He prepared us Long before even we identified him. My 6th President.

[Ashley Hunter @ashh640]

🚀 Introducing Angular Native

The stunning developer experience and ecosystem of Expo, using the Angular framework you already know and love.

• Real native iOS & Android
• Expo SDK & ecosystem
• Angular components, signals & DI
• CSS & Tailwind out of the box

https://t.co/U81C3Wuu16

[Angular @angular]

We’re rewriting the Angular compiler in Rust 🦀 (partially)!

Meet ngp (Angular Preprocessor): a hybrid compiler decoupling our transforms from TypeScript using Oxc, readying Angular for TypeScript 7 & native tooling.

Read the deep dive: https://t.co/d4a1EnlMdW 🛠️

[Kit Langton @kitlangton]

jesus one-shot christ

[Diogo Almeida @CompleteSkeptic]

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?

I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev

• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions

AFAICT the shortest path to AI-based economic revolution

[Chris Sev @chrissev]

the perfect skill doesnt exis...

/bro

[Kimi.ai @Kimi_Moonshot]

Releasing the model weights and technical report of Kimi K3.

Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window.

New model architecture: 2.5x the intelligence per unit of compute, not just more params.

Alongside Kimi K3, we're opening up more of the stack behind it — high-performance attention kernels, MoE communication library, and infrastructure for running agent environments at scale.

Model weights: https://t.co/7m7eEg6Y0B
Tech report: https://t.co/yeu6cjpMCT
Tech blog: https://t.co/YTfiMSNM1f

[san @saneord]

researcher: what is mitochondria?
claude to fbi:
https://t.co/Mw44YvCjaW

[Justine Moore @venturetwins]

Me using Claude Opus 4.8 to rename a file

[LangChain @LangChain]

https://t.co/3CgzH0p7tF

[OpenCode @opencode]

NVIDIA's new open source model is now free on OpenCode Zen

Nemotron 3 Super is a mid sized model that is
- fast
- fully open source
- 1M context

[dax @thdxr]

look markdown table rendering finally

[Frank Wang @fanjiewang]

Zen × MiniMax M2.5 - free for a week

wait what.. 2.5?!
yes sir.. feels like M2.1 just dropped yesterday
and now it’s already M2.5?!

better try it before M2.67 shows up next month