The purpose of this website it to gather important information and practical guides for systems powered by AMD Ryzen AI MAX and MAX+ processors. Our guides cover various topics and of course there is a separate section for the AI ones. You can also find some information on Strix Halo PCs and boards.
Strix Halo on Ubuntu looks great - Netstatz
The 2TB NVMe drive that came with this box is shown below. It was modified using gparted on the Ubuntu USB live boot prior to installation. New Linux users
Strix the Stateful Agent
Meet Strix. I built Strix initially just as a hack project, but it’s become a tremendous help. But also, it’s gotten a little weird at times. Strix is a stateful agent. An AI that remembers long after the conversation is finished.
Why Futhark?
A high-performance and high-level purely functional data-parallel array programming language that can execute on the GPU and CPU.
Modern GPU Programming For MLSys — Modern GPU Programming For MLSys
Machine learning systems sit at the heart of modern AI workloads. In these systems, performance often comes down to the quality of a small number of GPU kernels. Attention kernels, LLM prefill and decode kernels, low-precision block-scaled GEMMs, fused MoE layers, and other large fused kernels all directly shape end-to-end speed in both training and serving.
For LLMs and AI Systems — Cassidy Williams (cassidoo)
For LLMs and AI Systems — Cassidy Williams (cassidoo)
co/core · 5.0 ★ | at-store
co/core is a cooperative for AI inference — people pooling the Macs they already own to run open models for each other. Tag: Developer.
co/core — an AI cooperative
co/core is a cooperative for AI inference — people pooling the Macs they already own to run open models for each other, instead of renting from the big clouds. An experiment in AI infrastructure we build, share, and own together. Bring your existing OpenAI-compatible code, or share a Mac and help run it.
On-device intelligence for every product
We're building a frontier lab for on-device AI. Small, specialized models for audio, vision, and text, faster than the cloud and free.
Crossing the Hyper-Thread Boundary - exe.dev blog
Your VMs can share a CPU core. You get to decide who they share it with.
Async/await on the GPU
GPU code can now use Rust's async/await. We share the reasons why and what this unlocks for GPU programming.
M5Paper Color ESP32S3 Dev Kit (600x400, 4" SPECTRA 6 eInk Display)
PaperColor is a development board featuring a 4-inch E Ink Spectra 6 full-color e-paper display with a resolution of 400x600, offering both low power consumption and high visibility. It is built around an integrated ESP32-S3R8 core with 16MB Flash and 8MB PSRAM, and supports 2.4 GHz Wi-Fi wireless communication. The board incorporates a complete human-machine interaction system, including three programmable buttons, an audio system based on the ES8311 audio codec, a MEMS microphone with AEC echo cancellation, and a speaker driven by an AW8737A amplifier — enabling high-quality voice capture and audio playback. In terms of expanded functionality, the board integrates an SHT40 temperature and humidity sensor, RX8130CE real-time clock, microSD card slot, infrared emitter, two RGB LEDs, and an HY2.0-4P expansion interface. Combined with the M5PM1 multi-stage power management system and a 1250mAh battery, it enables long-lasting, low-power operation. Suitable for IoT and embedded applications such as electronic signage, voice interaction terminals, environmental monitoring instruments, and art display installations.
A new, modern UI component library built on top of Base UI - coss ui
Built for developers and AI.
Serpentine software
I don’t usually write hot takes. But StrongDM’s Software Factory post is worth it, so here goes. Their post is elegant, short, and well-written, so I’m going to just go through a bunch of pull quotes. Compounding correctness long-horizon agentic coding workflows [now] compound correctness rather than error This is a big, empirical claim. I’m already convinced by my own experiences that agents, suitably scaffolded, can form a Brownian ratchet, such that pouring on more compute helps, on balance.
I am often wrong
[I shared this note with my team earlier this week, and am posting it here as well. I hope it is interesting or helpful for others working on building product in the age of AI.]
Write-Only Code | Heavybit
AI may be accelerating software into a world of "write-only" code that forces a rethink of culture, processes, and tools. Heavybit’s Joseph Ruscio explains.
