Mar ia

Fullstack + DevOps in production | Next.js, AWS, Prisma | Writing about infra, auth & architecture | Learning out loud 🚀

[Jarred Sumner @jarredsumner]

I still want Bun to have a builtin on-disk fetch cache, for debugging & testing & avoiding rate limits in one-off scripts

maybe multiple backends, one with a folder of json files so that agents can debug with grep

[Freddy Montes @fmontes]

Cuando decide el mercado y no la empresa:

1. Mython es muy peligroso no lo podemos sacar
2. OpenAI asoma Sol
3. Creamos Fable y es "mas seguro" para ustedes
4. Usen Fable por 1 semana, luego paguen
5. OpenAI lanza Sol 2.5x mas barato
6. Usen Fable por 1 semana mas, luego paguen
7. Sale Kimi K3 muuuucho mas barato
8. Fable viene en tu plan

[Mar ia @maricarmen_dev]

Alerta de AWS Budgets: costo estimado de $11,026,587,972.99 USD.
Once mil millones. Revisé headers (no era phishing), revisé la config (no era eso), y al final: incidente global del Billing Console, "Inaccurate Estimated Billing Data".
Siempre revisa el status page primero.

[Miguel Ángel Durán @midudev]

¡Proyectazo! Necesitas esto para tu web o app.

Te genera automáticamente el skeleton de tus componentes. Sin calcular nada a mano.

Envuelves tus componentes con <Skeleton> y listo.

Para Vue, React, Angular y Svelte:
https://t.co/95p9i8FHrs

[shadcn @shadcn]

Introducing shadcn/typeset.

You know how you render markdown and get back plain, unstyled HTML? Headings, paragraphs, lists, tables. So you style the elements one by one: font sizes, line heights, spacing.

You do it for your blog. Then you do it again for docs. Then again for the chat. Every time, you're fighting the same thing: sizing and spacing.

To fix this, we created typeset.css: one file that styles everything inside a typeset container. It lives in your project, so you can change it directly when you need to.

And we made it work beautifully with streaming markdown.

[Guillermo Rauch @rauchg]

I love that @shadcn has managed to build the best UI infrastructure for agents and for humans. It's so so satisfying to play with the new typeset configurator.

[Jukan @jukan05]

CHINA CONSIDERS RESTRICTING OVERSEAS ACCESS TO CUTTING-EDGE AI MODELS

China’s Ministry of Commerce has led meetings over the past month with major AI companies, including Alibaba, ByteDance, and https://t.co/YDe0KRldDB, to discuss measures that would restrict overseas access to cutting-edge AI models, including models that have not yet been released.

The discussions reportedly include not only closed-source models but also open-weight models. However, the scope of application is still under debate, and the rules may ultimately apply only to future frontier models.

Officials have also discussed designating the leakage or theft of proprietary AI technologies as a national security crime, with stronger penalties, as well as restricting the types of foreign capital that can invest in Chinese AI startups.

The backdrop is the U.S. move to strengthen export controls on AI models, along with national security concerns over cutting-edge models that could possess advanced cyberattack capabilities.

Chinese authorities are reportedly concerned that advanced U.S. cybersecurity AI models could be used to exploit vulnerabilities in Chinese software.

Since the beginning of this year, China has continued to tighten measures to prevent AI technology from being transferred overseas. Authorities have investigated whether Chinese AI startups that relocated abroad violated export control laws, while also strengthening oversight of overseas transactions involving Chinese investors, technology, data, and national security concerns.

Future regulations could take the form of a tiered framework based on technological capability. Basic open-source AI models may be managed through a filing system, high-performance models may be subject to security reviews, and the most sensitive frontier models may be banned from public release or restricted to use within China.

[ClaudeDevs @ClaudeDevs]

6 months of Claude Max 20x, on us.

We're expanding Claude for Open Source to more of the community.

If you're a maintainer, a core contributor, someone landing PRs across the ecosystem, or someone keeping a critical package alive, apply today!

[Mar ia @maricarmen_dev]

¡Cosa bonita, cosa hermosa, cosa bien hecha! 😍

[Kamil Mysliwiec @kammysliwiec]

It's here. Our new @nestframework website is live. ✨ Our first major redesign in 9 years.

It's still us, just sharper - a cleaner, faster, and more modern experience built for the next chapter.

Take a look and let us know what you think!
https://t.co/WdbjnqGogD

[Kamil Mysliwiec @kammysliwiec]

It's here. Our new @nestframework website is live. ✨ Our first major redesign in 9 years.

It's still us, just sharper - a cleaner, faster, and more modern experience built for the next chapter.

Take a look and let us know what you think!
https://t.co/WdbjnqGogD

[santi @santtiagom_]

Anthropic publicó 5 workshops para aprender a construir agentes desde cero.

empiezan con un agente básico y, paso a paso, muestran cómo:

> agregar memoria
> hacerlo autónomo
> ejecutar tareas de forma proactiva
> construir agentes que mejoran con el tiempo

si están empezando con agentes, es un recurso que vale mucho la pena.

[Mar ia @maricarmen_dev]

Todos leen “Vercel compró Better Auth” como noticia de login.
La jugada real está al final del post: identidad de agentes. Permisos acotados y revocables cuando el agente actúa por ti.
Yo armo con Next + Prisma y era mi default. Ahora toca ver si sigue igual de agnóstico.

[Vercel @vercel]

The @better_auth team is joining Vercel to accelerate open source auth for apps and agents.

https://t.co/gboKZoHKz9

[Mar ia @maricarmen_dev]

@horbvert Es este https://t.co/z45QIahWBf
Super recomendado!

[Mar ia @maricarmen_dev]

Toca estudiar RAG y Embeddings

[Mar ia @maricarmen_dev]

Bonus para los que llegaron hasta acá 👇

AWS lanzó hace poco ECS Express Mode: le das tu imagen de contenedor y un solo comando te provisiona todo lo del diagrama — el service, el ALB con HTTPS, el auto scaling, el networking y hasta un dominio.

[Mar ia @maricarmen_dev]

¿Qué pasa entre tu docker push y el usuario final?

Amazon ECS en un diagrama, 8 pasos: build → ECR → task definition → service → scheduler → Fargate → ALB → CloudWatch.

[Mar ia @maricarmen_dev]

1. Empieza con Fargate. EC2 como launch type solo cuando el volumen justifique el ahorro.

2. El bug que más horas quema no es de red ni de config: es confundir el execution role con el task role. Uno arranca el contenedor, el otro es tu app.

[Mar ia @maricarmen_dev]

7/ No es moda, no es traición a ninguna herramienta. Es usar cada una para lo que de verdad es buena.

Generar código nunca fue el cuello de botella. Mantenerlo sí.

Análisis completo 👇
https://t.co/1stUr1FaIc

[Mar ia @maricarmen_dev]

1/ "¿Lovable y v0 ya pasaron de moda?"

No. Esa es la pregunta equivocada.

La buena es: ¿por qué decidí no meterlas a entregar producto? No es moda. Es esto 🧵

[Mar ia @maricarmen_dev]

6/ Por eso mi postura es clara:

✅ prototipar → sí, lo mejor que hay
❌ entregar producto real con ellas → no

El flujo que sigo: prototipo en Lovable/v0 → gradúo la lógica a Cursor o Claude Code cuando la idea demuestra que merece existir.