I just applied to speak at @RenderATL 2027. August 18–19 in Atlanta. Wish me luck! renderatl.com #RenderATL 🙌
1
41
JavaScript me ha dado muchas cosas geniales en la vida. Agrego una más: participar como speaker de un JSConf
Here's a recap (in Spanish 🇲🇽) of my experience at @vercel Ship 26.
More content coming in the next few days, as I put into practice everything I learned this past week. 🚀
1
97
In my last visit to San Francisco, I had the opportunity to know what the people at @vercel has been cooking. Great experience and tons of information.
1
274
Erik Porroa retweeted
I've joined @AnthropicAI!
I'll be working on Claude Code and making it better for developers who use it. Here's a little Fable & Three.js
Erik Porroa retweeted
It is probably the best Claude Design tutorial under 32 minutes
It is by UI Collective Academy
Erik Porroa retweeted
Anthropic a publié une Formation complet de 2 HEURES sur la construction d'agents Claude.
Animé par l'ingénieur qui construit Claude Code.
Gardez-la précieusement en Signet🔖
de A à Z : Structurer un agent qui se gère sans supervision. Lui donner accès au terminal pour exécuter, lire, corriger. Gérer sa mémoire via le système de fichiers. Bloquer les hallucinations avec des Hooks. Faire tourner un agent sur un gros codebase sans tout casser.
À la fin : vous utilisez Claude comme un pro et vous monétisez vos compétences. Débutant ou avancé, tout est là en un seul endroit, ce cours couvre tout.
Ça vaut plus que tous les cours à 500$ que t’as failli acheter.
Erik Porroa retweeted
INSTEAD OF WATCHING AN HOUR OF NETFLIX TONIGHT.
This 60-minute Cambridge lecture by Demis Hassabis will teach you more about the future of AI than most people will learn in the next 5 years.
Bookmark it and give it an hour, no matter what.
3 acts in 30 minutes: amnesia → memory → dreaming. 🧠
@soyrichlira showing how to build AI agents that never lose context with @claudeai at @platzi conf #PlatziConf
2
126
Great time I spent attending the Claude Code Meetup “en español” at @frontiertower. @AnthropicAI and @claudeai: Latino power has set a precedent today. Thanks @soyrichlira @cvander for bringing the party to SFO!
Erik Porroa retweeted
🚨Claude just broke the AI industry. Again.
This time it's sitting at your desk, running your apps, doing your work.
Text it from your phone. Come back to finished work. That's literally it.
Here's what it can do (🔖 Save this cheatsheet):
→ Open apps, find files, export, attach, send
→ Fill spreadsheets from browser tabs
→ Schedule emails and morning briefings
→ Submit PRs and run tests
→ Organize your Downloads folder
→ Update your calendar from Slack messages
Erik Porroa retweeted
Someone quietly gave Claude a superpower… and it’s wild.
It now has its own browser.
Not controlled. Not supervised.
Independent.
→ Logs into accounts
→ Scrapes live data
→ Watches communities
→ Pulls leads automatically
And it runs… even when they’re offline.
No prompts every step.
No manual work.
They just connected Firecrawl once—
now Claude executes tasks like:
“Visit this site → extract info → organize it”
And it delivers clean results.
This isn’t a chatbot anymore.
It’s a background operator.
Most people haven’t realized this yet
Erik Porroa retweeted
THE HERMES AGENT IS GOING VIRAL RIGHT NOW. HERE'S WHY PEOPLE ARE CALLING IT THE NEXT OPENCLAW:
someone made a full breakdown comparing them side by side.
openclaw: 339K github stars. 5,400+ skills. massive ecosystem. you plug in tools and let them execute.
hermes: 14.6K stars. smaller. but it does something openclaw doesn't.
it learns from experience. creates its own skills. improves them over time. and builds a model of YOU across sessions.
openclaw is config-first. write a SOUL.md, connect your channels, go.
hermes is learning-loop-first. the agent creates skills from experience, improves them, and deepens its model of you the more you use it.
memory:
> openclaw: session-based. persistence through plugins
> hermes: async writes, configurable recall, multi-user isolation
self-improvement:
> openclaw: no built-in learning loop. skills installed manually
> hermes: core differentiator. creates skills from experience and improves them automatically
security:
> openclaw: 36% of ClawHub skills contain prompt injections. 135K+ unprotected instances
> hermes: smart approvals, kill switch, filesystem snapshots before destructive ops
the tldr:
openclaw does the work. hermes does the thinking.
openclaw is for tool runners who want a massive ecosystem. hermes is for people who want a meta-agent that gets smarter over time.
both are open source and both are free but the better one depends on what you're trying to do
Erik Porroa retweeted
Learn algorithms - visually!
An excellent collection of interactive algorithms organized by category. Check it out.
algorithm-visualizer.org/
Hoy iniciamos la Claude Code Meetup en Monterrey, una oportunidad para conocer más a profundidad la herramienta de @AnthropicAI y encontrarnos con otros creadores en la ciudad. ¡Gracias a @soyrichlira por traer el tour de @claudeai por aquí!
2
86
Erik Porroa retweeted
Conoce a los ponentes que serán parte de #JSConfMX 2025🤩
Nos compartirán de buenas prácticas, demos, conocimiento y aprendizajes. Además, podrás convivir con ellos durante la conf🔥
¡No te quedes fuera!
Aparta tu lugar en 👉jsconf.mx
#JSConf #Javascript #Community






























