Week 2 Recap: the Protocol Builders Program cohort went deeper into the technical foundations of protocol engineering on @Polkadot. Advanced sessions, complex architectural concepts, real-world protocol design, and focused collaboration with distinguished faculty. The atmosphere was focused, rigorous, and collaborative, with students and faculty working side by side to turn technical knowledge into deeper understanding.

¡Hoy es un día histórico para JavaScript! Temporal API ya disponible en Node.js. Por fin una API decente para fechas. Adiós a new Date("2026-04-28") Hola a Temporal.PlainDate.from("2026-04-28") ✓ Fechas sin zonas horarias raras ✓ Suma días, meses y años sin problemas ✓ Adiós a muchos bugs de Date

LambdaClass is launching an Engineering Residency in Buenos Aires for people who want to work on System programming, Distributed Systems, Cryptography, High-performance, low-level infrastructure. This is a 6-month, full-time, on-site opportunity, working on real systems and applied research. Before joining, candidates must complete our Hacking Learning Path autonomously.

Anthropic paga más de 750.000 $ al año para entender los LLMs. Stanford ha publicado una conferencia de 2 horas que cubre el 80% gratis. Puede ser lo más rentable que hagas este mes. (Guárdalo, te servirá)

Google descubrió un truco muy estúpido para que la inteligencia artificial te responda MUCHO mejor. Mientras todo el mundo busca el prompt perfecto, escribe instrucciones complejas y técnicas con nombres elegantes... unos investigadores de Google descubrieron que lo que mejor funciona es copiar y pegar tu pregunta dos veces.  ¡Sí, así de simple! Le mandás el mismo texto dos veces seguidas y la IA responde mejor. Probaron con Gemini, ChatGPT, Claude y DeepSeek: en matemáticas, razonamiento y comprensión lectora mejoró en todo. En un caso, un modelo pasó de 21% de precisión a 97%. ¿Y por qué carajo funciona esto? Porque la IA lee tu mensaje una sola vez, de izquierda a derecha, sin poder volver atrás. Si tu pregunta está al final, todo el contexto de antes se procesó sin saber qué le ibas a preguntar. Es como leer un capítulo entero de un libro y recién al final alguien te dice: "ah, buscá cuántas veces aparece la palabra perro". Ya es tarde. Pero si lo leés dos veces, la segunda vez ya sabés qué buscar. Mientras vos buscabas el prompt perfecto, la respuesta era Ctrl C, Ctrl V. A veces la solución más bruta es la más inteligente.

Super happy and excited to be part of @AcademyPolkadot in this edition of PBP. Even more excited after seeing the profiles of the people I’ll be sharing these two weeks of learning with.

¡Esta funcionalidad de Claude Code es una locura! Crea un equipo de agentes en un minuto:

Los mejores repositorios de GitHub para Claude Code que multiplicarán por 10 tu próximo proyecto en 2026: 1. Claude Mem : github.com/thedotmack/claude… 2. UI UX Pro Max : github.com/nextlevelbuilder/… 3. n8n-MCP : github.com/czlonkowski/n8n-m… 4. Obsidian Skills : github.com/kepano/obsidian-s… 5. LightRAG : github.com/hkuds/lightrag 6. Everything Claude Code : github.com/affaan-m/everythi… 7. Superpowers : github.com/obra/superpowers 8. Awesome Claude Code : github.com/hesreallyhim/awes… 9. GSD (Get Shit Done) : github.com/gsd-build/get-shi…

Stop burning tokens on Claude Code. Use this instead 👇 A free GitHub repo (80K⭐) that turns your CLI into a high-performance AI coding system. Link → github.com/affaan-m/everythi… Why it’s different: → Token optimization Smart model selection lean prompts = lower cost → Memory persistence Auto-save/load context across sessions (No more losing the thread) → Continuous learning Turns your past work into reusable skills → Verification loops Built-in evals to make sure code actually works → Subagent orchestration Handles large codebases with iterative retrieval Most people think Claude struggles with complex repos. It doesn’t. They’re just not using the right setup. This fixes that. Bookmark this for your AI stack. ♻️ #AI #Claude #AIAgents #LLM #GenAI #DevTools

🚨Alibaba lanza CoPaw, competencia simple para Claude Cowork (pero gratis), guarda esto 💾 Básicamente un agente de IA personal/open source, que: ✅ Vive en tu compu (Ollama Qwen gratis) ✅ Tiene memoria real — no olvida nada ✅ Ejecuta tareas solo, con cron ✅ Podés programarle skills custom ✅ Cero pesos al mes Bienvenidos a la era del asistente propio. No el de la nube de otro — el tuyo. GitHub: github.com/agentscope-ai/CoP…

I was a 10x engineer. Now I'm useless.

We’re acquiring Promptfoo. Their technology will strengthen agentic security testing and evaluation capabilities in OpenAI Frontier. Promptfoo will remain open source under the current license, and we will continue to service and support current customers. openai.com/index/openai-to-a…

RT @codspeedhq: Stop Chasing Noise, Find the Signal.

Let me go deeper on what's happening under the hood here The pipeline has four stages: INPUT: Your secrets (amount, a secret key, a threshold) go into a Noir circuit. The circuit compiles from .nr to .json PROVE: Barretenberg takes that compiled circuit and generates an 8KB proof blob. This is the heavy part, 30-60 seconds. The circuit witness gets solved, then proven. What comes out is a cryptographic guarantee that your inputs satisfy the constraints without exposing them CONVERT: Garaga turns that proof into calldata (a felt[] array) that a Cairo contract can consume. This is the bridge between off-chain proving and on-chain verification VERIFY: A Starknet contract checks the proof. If it passes, a badge gets minted. Onchain, nobody sees the amount or the recipient. They just see that the proof is valid

Starknet is scaling Ethereum with ZK technology. But the biggest unlock isn't the tech. It's the builders. That's why we built Starknet Academy to turn curious devs into confident onchain builders. Here's what you get: ⚫️ Structured courses from beginner to advanced ⚫️ Hands-on Cairo programming ⚫️ Deep dives into ZK proofs & Starknet architecture ⚫️ Real projects to ship onchain All free. All in one place. 100% self paced. Visit academy.starknet.org

leverage like 2007, the golden rush are based on $20 AI subscriptions * 1 billion paid users. Imagine what happened after these people lose a job? Be prepared for AI margin call

T-minus 2h30m Klivi hasn’t blinked in hours. He doesn’t know if he’s debugging or hallucinating anymore. But somehow… he believes. #StarkNet #Hackathon #ZK @Starknet @StarknetFndn @CairoLang @StarknetEco @aztecnetwork Caffeine: 120%. Sanity: 3%. Hope: infinite.

wow

The rules of the game are beginning to change. Ready or not, ZK will be everywhere.

That's really fantastic. The new era of ZK is here :D