I built an open source TPU chip in 4 days 8x8 systolic array, 16 instruction ISA, 64 MACs/cycle all in ~3,400 lines of SystemVerilog I sent it to Tiny Tapeout for fabrication soon here's the full arch

Microsoft just solved the "Agent Loop" problem 🤯 They've open-sourced Agent Lightning, a framework that lets agents learn from their own mistakes. It uses Reinforcement Learning to optimize prompts and weights automatically. → Your agent tries a task → It fails → Agent

First stable release of RightNow

RightNow v1.0.0 Agents, skills, and expanded GPU language support This release focuses on extensibility and stability for GPU and systems development What changed: - Custom agents can be defined and used directly in chat, including DSLs focused agents - Agents can be extended

You don't need an H100 to test on one every NVIDIA GPU architecture has a publicly documented instruction set, register file layout, memory hierarchy, and scheduling model. All of it is out there so we built an emulator that models 86+ architectures with under 2% error. Same

i built a local LLM inference engine that runs a 1B parameter model on a $10 board with 256mb ram. model sits on the sd card, streams one layer at a time through 45mb of ram You can use it as local LLM model backend for PicoClaw no python no cloud no api keys 80kb binary | pure

api.rightnowai.co

We open sourced an operating system for ai agents 137k lines of rust, MIT licensed we love OpenClaw🦞 and it inspired a lot of what we built. but we wanted something that works at the kernel level so we built OpenFang agents run inside WASM sandboxes the same way processes