Aral Roca

Let's go! 🔥🔥 Brisa Framework can now be used! 😁

Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI

🦀 Microsoft just open-sourced a comprehensive Rust training curriculum and it's impressive. The microsoft/RustTraining repository on GitHub offers 7 structured books covering Rust from beginner to expert level, designed for developers coming from different backgrounds: 🟢 Bridge Books (start here): • Rust for C/C++ Programmers • Rust for C# Programmers • Rust for Python Programmers 🔵 Deep Dive: Async Rust (Tokio, streams, cancellation) 🟡 Advanced: Rust Patterns (Pin, allocators, lock-free structures) 🟣 Expert: Type-Driven Correctness (type-state, phantom types) 🟤 Practices: Rust Engineering (CI/CD, cross-compilation, Miri) Each book includes 15–16 chapters, Mermaid diagrams, interactive Rust playgrounds, and exercises. Whether you're a systems programmer migrating from C++, a .NET developer exploring performance-critical code, or a Pythonista tired of the GIL, there's a path for you. ⭐ Already at 500+ stars. Fully open source (MIT + CC-BY-SA-4.0). 👉 github.com/microsoft/Rust… Rust is becoming a serious part of the industry stack. If you've been waiting for a structured way to learn it, this might be it. #Rust #Programming #OpenSource #Microsoft #SoftwareEngineering #SystemsProgramming #Learning

Introducing Hyperagents: an AI system that not only improves at solving tasks, but also improves how it improves itself. The Darwin Gödel Machine (DGM) demonstrated that open-ended self-improvement is possible by iteratively generating and evaluating improved agents, yet it relies on a key assumption: that improvements in task performance (e.g., coding ability) translate into improvements in the self-improvement process itself. This alignment holds in coding, where both evaluation and modification are expressed in the same domain, but breaks down more generally. As a result, prior systems remain constrained by fixed, handcrafted meta-level procedures that do not themselves evolve. We introduce Hyperagents – self-referential agents that can modify both their task-solving behavior and the process that generates future improvements. This enables what we call metacognitive self-modification: learning not just to perform better, but to improve at improving. We instantiate this framework as DGM-Hyperagents (DGM-H), an extension of the DGM in which both task-solving behavior and the self-improvement procedure are editable and subject to evolution. Across diverse domains (coding, paper review, robotics reward design, and Olympiad-level math solution grading), hyperagents enable continuous performance improvements over time and outperform baselines without self-improvement or open-ended exploration, as well as prior self-improving systems (including DGM). DGM-H also improves the process by which new agents are generated (e.g. persistent memory, performance tracking), and these meta-level improvements transfer across domains and accumulate across runs. This work was done during my internship at Meta (@AIatMeta), in collaboration with Bingchen Zhao (@BingchenZhao), Wannan Yang (@winnieyangwn), Jakob Foerster (@j_foerst), Jeff Clune (@jeffclune), Minqi Jiang (@MinqiJiang), Sam Devlin (@smdvln), and Tatiana Shavrina (@rybolos).

BTW: the one in the photo is my mom 🤣 All this soon in kitmul.com/en

Meanwhile, in my free time, I’m building an AI that orchestrates other AIs… while fixing bugs in my libraries (next-translate & teaful) 😂 It’s still a work in progress (spoiler)… so far I’ve created 206 tools, but the goal is 4,000+ I’m having a lot of fun with it 💪😂

Teaful 0.12 is here 💪🏽 A brand new implementation powered by useSyncExternalStore. Now even tinier: just ~800B!

Next-Translate 3.1 is here 💪🏽 github.com/aralroca/next-…

Read this over the weekend (bonus if you read all the papers in the Research List) and you’ll be among the “very few who understand how far-reaching” the shift to World Models is. notboring.co/p/world-models

@richterich_ @ThePrimeagen no login, 100% local kitmul.com/en/image-desig…

@ThePrimeagen Try converting from webp without a "free trial" or a credit card

why does everyone hate webp?

@galigan I utilitzant open.maic.chat pot crear les classes interactives per aprendre

Saps què hauria d'aprendre el teu fill a 2n d'ESO? Jo tampoc ho sabia. Ara sí: he agafat el currículum oficial de la Generalitat i l'he convertit en una web on pots consultar-ho tot en dos clics. 👉 curriculumsecundaria.vercel.app

Con IA estoy construyendo una IA capaz de orquestar otras IAs. Muy pronto en kitmul.com/es Empecé creándolo con Next.js para mantener y experimentar con next-translate y Teaful (las librerías que mantengo). Ha evolucionado en un side-project muy emocionante ☺️

In the next version of Bun `Bun.WebView` programmatically controls a headless web browser in Bun

🚨 Ya está abierta la convocatoria del Impact Challenge de @Googleorg: IA para la Ciencia. Un fondo global de 30 millones de USD para acelerar descubrimientos científicos en salud y resiliencia climática. Qué reciben los proyectos seleccionados: ✅ Entre $500,000 y $3 millones de USD. ✅ Acceso a la aceleradora de Google.org. ✅ Mentoría técnica e infraestructura en la nube. Postulaciones abiertas hasta el 17 de abril de 2026. Aplica aquí 👉 google.org/impact-challen…

Want to learn? - OpenMAIC: Generate interactive lessons - LLMNotebook: Generate podcasts based on content - ElevenReader: Generate audiobooks from EPUB & PDF files Use AI to boost your brainpower.

This is just AMAZING 🤯 github.com/THU-MAIC/OpenM…

AI model to WASM + Worker! dev.to/aralroca/i-bui…

GLM-OCR runs locally on 2GB VRAM, handles tables and math equations, and hits 260 tok/s on a Mac. No cloud API. No subscription. Just your machine. Local models are getting better and smaller faster than anyone expected.

.@nodejs has always been about I/O. Streams, buffers, sockets, files. But there's a gap that has bugged me for years: you can't virtualize the filesystem. You can't import a module that only exists in memory. You can't bundle assets into a Single Executable without patching half the standard library. That changes now 👇

I built my own Pomodoro Timer with radio music included, even Lofi Girl 🍅 I’m done relying on external tools. Time to build my own. kitmul.com/en/agile-proje…