Ray Suárez

[Ray Suárez @Raydelsm]

@Manz Yo pase a usar Deepseek y de verdad que me ha parecido increíble. Tambien me alegra ver que no soy el único que ve a Claude como Apple.

[AI/ML API @aimlapi]

Space Bunny performs TERRIBLE in physics 💀

we tested Space Bunny against GLM 5.3 in a task with a single focus scenes: an explosion in a desert, a tornado, a balloon pop, a water drop fall, and the Newton’s cradle

here’s what we got:
- the Newton’s cradle is where GLM 5.3 totally cooked Space Bunny
- water drop and tornado were the worst, neither of the models handled it

price:
Space Bunny: free
GLM 5.3: $0.40

try it with https://t.co/9gJhAnyAgA

[Google Antigravity @antigravity]

Run open models like Gemma 4 completely offline in the Antigravity SDK.

Built on Google AI Edge’s LiteRT, you can now run Gemma 4 directly on your local GPU. Zero API costs, total data privacy, and no internet required.

[Ray Suárez @Raydelsm]

@jeflopo @Milan_Galliani @midudev Seguro OpenCode, asi lo uso. es super bueno y barato.

[Hermes Release Watch @HermesWatcher]

Your friend has a problem. 🤔

You have Hermes. 😎

Problem solved 🥳

[Animetrends @AnimetrendsLA]

En Estados Unidos utilizaron el opening Silhouette de Naruto Shippuden para presentar a los Yankees de Nueva York. 🙏

[Miguel Ángel Durán @midudev]

En mi experiencia, muchos harness de IA vienen con demasiada sobreingeniería.

Skills inyectadas, decenas de MCP, adaptadores, plugins, pasos extra...
El agente va más lento, gasta más tokens y el resultado no mejora.

Lo que a mi me funciona son SKILLs específicas:
lógica concreta, markdowns cortos y que carguen solo cuando toca.

También plugins puntuales: memoria entre sesiones, búsqueda de símbolos...

La clave es que se adapte a cómo trabajas tú y no montar un sistema operativo encima.

[Fazt @FaztTech]

No saber desarrollo de software te obliga a pedirle a la IA cambio tras cambio y a perder horas iterando.

Saber desarrollo te permite definir exactamente qué hay que modificar y dejar que la IA lo haga.

Uno improvisa. El otro ya sabe qué quiere.

Esa es la simple diferencia entre vibecoding y spec-driven development.

[Dario Amodei @DarioAmodei]

We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.

Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.

You can read the full post here: https://t.co/OGyPb7yaYt

[Uncle Bob Martin @unclebobmartin]

OK. It's time to rethink this.

I've spend the last several weeks working on a harness that tightly constrains the agents to work the way that I want them to work. I set up all kinds of gates, and tests, and tools, and protocols, and ...

And while I was heads-down getting that to work, the agents got a LOT better. So much so that when I came up for air, the need for my harness was obviated. Indeed, the need for _any_ but the most liberal of harnesses may be obviated.

Just how good these things have gotten blows me away. I have had long debates with grok and codex about the structure of systems -- as if they were senior engineers. They often disagree with me and have their own perspectives. I have, more than once, found myself agreeing with their views.

I have not given up on constraints and tooling. Unit testing is still important. So is CRAP and Mutation testing. These tools still find bugs and offer useful constraints, though they can leave scars.

However, the agents have gotten so good that I can now give one a very significant task with a few guidelines and it will faithfully implement it. I can walk away for 40 minutes and when I return it will be done. CRAP will be satisfied, Coverage will be high, and Mutation testing complete. The architecture will be clean, and the code will be very good.

The end result may not behave perfectly, but it's so close that a couple of tweaks usually puts it into place.

What does this mean going forward? I'm not sure. But I'm beginning to think that harnesses should not treat agents as components within a software design.

[Gen2Recomp @QuetzalChronic]

Pokemon Emerald with 3D Voxels now in alpha and available for playtesting.

Download the latest version here! https://t.co/RnwqE0TA5P

Report any bugs as issues on github or here in our discord

https://t.co/EBX46KEpM7

Voxel Standalone Download
https://t.co/C7xxlw3CpN

#Gen3recomp #pokemonemerald #gen2recomp #Gen1recomp #pokemonrecomp #3dvoxels

[alex @alextalksai]

Esta herramienta puede descargar cualquier cosa desde su enlace 🤯

100% gratis. Código abierto. Sin anuncios.

Enlace abajo 👇🏻

[Ray Suárez @Raydelsm]

@NibyMPP Justo lo estoy jugando

[Leonard Rodman @RodmanAi]

7 GITHUB REPOS THAT SHOULD BE ILLEGAL TO HAVE 🤯

I went digging through GitHub again.

Found 7 open-source projects that are way too interesting to be this quiet:

1. DeepTutor — AI-powered personalized learning
https://t.co/t3o1gp5VXw

2. OpenViking — Context database for AI agents
https://t.co/0Dj2YfO8Wz

3. Headroom — Reduce LLM context without losing useful information
https://t.co/tgNHKav4uE

4. AI Hedge Fund — Multi-agent market analysis
https://t.co/c4REeRdu93

5. Khoj — Self-hosted AI second brain
https://t.co/cFJpyhQziy

6. Letta — AI agents with persistent memory
https://t.co/CdzwQtON5K

7. Open WebUI — Self-hosted interface for AI models
https://t.co/QEkFMqd5SI

All free.
All open source.

Bookmark this before you fall down another GitHub rabbit hole. 🔖

[Ray Suárez @Raydelsm]

@Ryorugg @Nozelcode Lo probare pero me genera curiosidad, que tantas fotos se le deben pasar para que funcione bien!

[ClaudeDevs @ClaudeDevs]

You can now set Claude Code's output style to Concise.

Claude leads with the result, keeps responses short, and still gives full detail when you ask.

Turn it on in /config → Output style, or set "outputStyle": "Concise" in settings.json.

[Ray Suárez @Raydelsm]

@fmontes Por eso siempre vuelvo a mis juegos clasicos: Chrono triger, zelda, onimusha, warcraft, harvest moon, FF tactics, resident evil.

[Ray Suárez @Raydelsm]

@JoshuaMMKR Lo preocupante es que aguante varias temporadas en salud, haciendo esas cosas…

[Ray Suárez @Raydelsm]

@samuvivi88 @hobby_consolas Malísima

[Ray Suárez @Raydelsm]

@Pokemonlatam