_$udu$_ retweeted
POV: Antrhopic’s social media intern posting the new Claude update of the day.
_$udu$_ retweeted
you mean this free doc?
A senior Google engineer just dropped a 421-page doc called Agentic Design Patterns.
Every chapter is code-backed and covers the frontier of AI systems:
→ Prompt chaining, routing, memory
→ MCP & multi-agent coordination
→ Guardrails, reasoning, planning
This isn’t a blog post. It’s a curriculum. And it’s free.
_$udu$_ retweeted
people who've been grinding in silence for 3 years after reading the "let them" theory for the first time
_$udu$_ retweeted
"It's not [fans'] job to keep theaters open, it's our job to make things that make it worth coming out"
— 'Project Hail Mary' star Ryan Gosling
_$udu$_ retweeted
🚨 Do you understand what Apple just did..
the company that built its entire brand on "we don't sell ads.. we sell products".. is now putting advertising inside Apple Maps..
Steve Jobs stood on stage and said the iPhone was about the user experience.. no distractions.. "it just works"..
now you're going to get a McDonald's ad while you're trying to not miss your exit..
Apple charged you $1,200 for the phone.. $100/year for iCloud.. $10/month for Apple Music.. $7/month for Apple TV.. and after all of that.. they still want ad money..
this is the same company that ran an entire ad campaign attacking Google for tracking you.. "privacy.. that's iPhone".. remember that?..
turns out privacy was just the sales pitch until they figured out how to sell you themselves.
_$udu$_ retweeted
Claude can now control your entire computer with one prompt. forever. read that again.
you tell it “scan my email every morning” once. it just does it, forever.
it can also open apps, edit spreadsheets, move files, batch process 150 photos in Photoshop, export PDFs. all of it.
you can also start a task from your phone, go to dinner, come back to finished work like nothing happened.
anthropic is giving us the full desktop agent that uses the actual screen, mouse, and keyboard. not a sandbox. not a simulation. your real jarvis.
this is absolutely insane. the LLM era is over.
_$udu$_ retweeted
How to setup your Claude code project?
TL;DR
Most developers skip the setup and just start prompting. That's the mistake.
A proper Claude Code project lives inside a .𝗰𝗹𝗮𝘂𝗱𝗲/ folder. Start with 𝗖𝗟𝗔𝗨𝗗𝗘.𝗺𝗱 as Claude's instruction manual. Split it into a 𝗿𝘂𝗹𝗲𝘀/ folder as it grows. Add 𝗰𝗼𝗺𝗺𝗮𝗻𝗱𝘀/ for repeatable workflows, 𝘀𝗸𝗶𝗹𝗹𝘀/ for context-triggered automation, and 𝗮𝗴𝗲𝗻𝘁𝘀/ for isolated subagents. Lock down permissions in 𝘀𝗲𝘁𝘁𝗶𝗻𝗴𝘀.𝗷𝘀𝗼𝗻.
There are two .𝗰𝗹𝗮𝘂𝗱𝗲/ folders: one committed with your repo, one global at ~/.𝗰𝗹𝗮𝘂𝗱𝗲/ for personal preferences and auto-memory across projects.
The .𝗰𝗹𝗮𝘂𝗱𝗲/ folder is infrastructure. Treat it like one.
The article below is a complete guide to 𝗖𝗟𝗔𝗨𝗗𝗘.𝗺𝗱, custom commands, skills, agents, and permissions, and how to set them up properly.
_$udu$_ retweeted
La pregunta no es por qué El Salvador se ha levantado tan rápido sino por qué Europa se está cayendo tan rápido. Recuerda, no hay Gobierno que no pueda combatir la delincuencia. Si no la combate es porque es cómplice. Así de sencillo.
_$udu$_ retweeted
ok but did he get a job
THIS GUY AUTOMATED HIS ENTIRE JOB SEARCH WITH AI AGENTS
instead of applying manually
he built a system with Claude Code that scores every job offer against 10 filters and auto-applies to the ones that match.
the agents:
> Scanner: discovers listings across 20+ job portals
> Evaluator: scores each job A-F against 10 dimensions
> PDF Generator: creates an ATS-optimized resume customized per offer
> Applicator: fills out application forms automatically via Playwright
the 10 filters:
> role fit (does the job match one of his 6 target role types)
> skills alignment vs. proof points
> seniority calibration
> compensation range
> geographic viability
> company stability
> product-market interest
> growth trajectory
> ATS compatibility
> timeline urgency
results: 516 jobs evaluated. 66 applications sent. 450 auto-rejected with documented reasoning so that the model gets even better.
the whole system runs as 12 Claude Code skills. each with its own context doc.
the system itself demonstrates exactly the skills the target roles require. multi-agent orchestration, production automation, human-in-the-loop design.
this is what job searching looks like in 2026.
_$udu$_ retweeted
Reminds me to that joke about USA seeing pens not working in space so they spend a billion to create a pen that works in space. Soviets brought a pencil.
Assyrians raised and killed sheep, worked the skin, sew the leather, added an inflater; and Greeks were like: uhm wood floats
_$udu$_ retweeted
LLMs starting to evolve defenses to distillation attacks.
Interesting.
