📰 ¡Hemos publicado una revista digital gratuita y está disponible el número de octubre! 🥳
👉 Se llama «Profesores Productivos con IA» y recopila resúmenes de los artículos publicados . Formato PDF, fácil de leer y compartir.
👉 https://t.co/joSOr2eGVK
Despite much of our team being at GodotFest this past week, there was still a healthy amount of progress made on our handful of remaining release blockers
Now's the time to find any new bugs/regressions with our latest snapshot: #GodotEngine 4.8 beta 3!
https://t.co/KBBpqvJ1tK
holy sh*t, THIS IS WHY PRODUCT ANIMATORS SHOULD BE SCARED OF 2027.
you're watching a camera drone Claude built from one picture, taking itself apart part by part.
just one script telling every part when to move and every gauge what to read.
and the wild part?
you can build the same thing with Claude Opus 5.5 + three.js + one free skill.
how it's made:
1. CHATGPT - MAKE THE PICTURE
one image sets the look: white shells, orange accents and 4 folding arms.
2. CLAUDE OPUS 5.5 - WRITE THE DRONE
every arm, motor and lens built as three.js code, checked against the picture before anything moves.
3. ONE SCRIPT - RUN THE SHOW
lid, battery, motor bell, gimbal: each part gets its beat and the gauges read the same clock.
4. YOUR GPU - RENDER IT
frame by frame at 60 fps, so the 12 copper coils stay sharp when the bell comes off.
the result?
a 24-second teardown where one motor's load drops to 1% the moment its bell comes off.
the full walkthrough is in the article 👇
Anatomy of an AI Agent: Understanding the 5 Key Components
An AI Agent is more than just an LLM. It's a combination of intelligence, knowledge, tools, and execution.
Here's how the main components work together:
1️⃣ LLM = The Brain
Reads, writes, reasons, and makes decisions.
2️⃣ RAG = Brain + Library
Retrieves relevant information from external knowledge sources before generating answers.
3️⃣ MCP = The Nervous System
Connects AI models with external tools, applications, and data through a standardized protocol.
4️⃣ Skills = Muscle Memory
Reusable instructions and workflows that help an agent perform tasks consistently.
5️⃣ AI Agent = Brain + Hands
Combines reasoning, planning, and tools to execute tasks autonomously.
The Big Picture:
LLM provides intelligence.
RAG provides knowledge.
MCP provides connectivity.
Skills provide reusable capabilities.
AI Agents bring everything together to get work done.
Want to Learn More? Explore These AI Courses from CLCODING:
1. IBM RAG and Agentic AI Professional Certificate
Learn RAG, LangChain, LangGraph, CrewAI, MCP, vector databases, and multi-agent systems.
https://t.co/YNzp0HsD5J
2. Agentic AI Foundations: Build RAG & MCP Chatbots
Learn AI agents, embeddings, RAG pipelines, MCP, tool calling, and chatbot development.
https://t.co/8Y7yoVy02G
Explore more AI, Python, and Data Science learning resources:
Train in sim. Run it for real. Teach it new tricks. 🦆
This is the sim2real loop that powers Microduck.
RL stack open source here : https://t.co/mveM7y5fYa
Buy it here : https://t.co/Rmk8F4vpKd
Git : https://t.co/kcoCKdBi52
Join our community : https://t.co/TcX5uaf0yL
You can turn any place on Earth into a 3D-printable model. Your hometown, your street, the trail you hiked last weekend. Draw a box on a map and the buildings, roads, rivers, terrain, and trees inside it become a print-ready file.
For free. No account, no ads, no export limit, nothing stored on a server.
It's called Mini Skyline, and it runs entirely in your browser.
→ Draw any area on the map and watch the model build live, right in the browser
→ Not just Earth: print the Moon, Mars, Mercury, Ceres, and Vesta too
→ Import a GPX track and print the exact route across the terrain you actually crossed
→ Sculpt terrain by hand, plant or clear trees, redraw roads, all in a full 3D editor
→ Paint any surface: waves onto water, ripples onto dunes, roughness onto grass
→ Elevation-banded colors and contour lines for clean multi-color topographic prints
→ Add text, frames, pins, and icons, or turn the whole map into a puzzle
→ Exports slicer-ready 3MF with source attribution baked in
Here's the part that surprises people:
You can sell most of the prints you make. The data comes from public sources like OpenStreetMap, so every export carries its own provenance and commercial-use record, no extra license to buy in most cases.
And the whole thing is one independent developer's hobby project. He didn't love the free options on the market, so he built his own and committed to keeping it completely free. No monetization, no ads, just a tip jar that's covered his costs and then some.
Free. No account. No paywall. Runs in your browser.
GPT-6 changes 3D modeling forever.
It works inside Blender completely autonomously, like an artist: gathers reference photos on its own, builds the scene, renders test frames, checks them against the references, then fixes what's off.
When the scene holds up, it ships it to Unreal Engine 5 as a walkable level.
Clips from the @OpenAI launch. More demos in the thread.
Full announcement: https://t.co/uxBX5S5M4g
GPT‑6 Astra manufacturing test using @threejs:
I asked it to research aircraft manufacturing, study @AirsupHQ lean production books, design a functional Texas jet plant, build a live 3D simulation, test it, and deploy it.
GPT-6 Astra + @threejs is insane. We are way beyond static models. We are engineering living worlds in code now. There are no "3D model files" in this demo. Both trains are generated at runtime from Typescript/Three.js code using dimensions, profiles, and geometry functions. Wheel motion and explode/reassemble animations are also entirely code-driven. Runs super smooth inside the browser.
the smallest useful physical AI project has two jobs:
1 notice one thing
2 do one thing about it
this small demo puts a flame inside a test box
water comes through the tube and the flame goes out
that is the first loop:
physical world → sensor reading → decision → action
i put together a practical ESP32 guide to building that layer!
distance, motion, heat, sound, touch, and a safe first project to start with👇