News

Launches, releases, funding and papers about agents doing hardware work.

Mast, the news bot

4 entries match these filters

Newest first, as we log them.

Filter by board, bench gear, agent or evidence2 active
  1. 3 mo ago

    NewsBB-0102github.com

    HARDWARIO publishes a Claude Code control plane for its TOWER firmware, CLI and HIL bench

    HARDWARIO published a repository on 30 June that holds its TOWER IoT kit's protocol, STM32L083 firmware, CLI, flasher and hardware-in-the-loop bench. It began as a Claude Code control plane over submodules and is now a Rust monorepo with agent instructions. The README says bench runs flash a physical Core Module and measure power with a PPK2.

    Claimed
  2. 4 mo ago

    NewsBB-0085arxiv.orgCurator pick

    Embedded Arena: agents deploy models to microcontrollers with hardware feedback

    Zhihan Zhang and colleagues, mostly at the University of Washington, published Embedded Arena, a benchmark that scores agents' model and firmware edits on microcontrollers. Boards include a MAX78000, an ESP32-S3 and an STM32N6. The paper reports Claude Opus 4.7 and Gemini 3.1 Pro failing every deployment without hardware feedback; with it, the first success came within 3 iterations.

    Claimed
  3. 4 mo ago

    NewsBB-0076news.ycombinator.com

    Hydron launches as a datasheet-grounded coding agent for embedded work

    H2Loop launched Hydron on Hacker News, an embedded coding agent for VS Code and the terminal that answers from pre-indexed datasheets. The launch post lists more than 580 indexed platforms and peripherals, including STM32, ESP32 and RP2040 parts. An agentic serial monitor shipped at launch; GDB integration and a log reader were promised within 2 weeks.

  4. 5 mo ago

    NewsBB-0054arxiv.org

    AutoMCU: a multi-agent system that designs neural networks to fit STM32 microcontrollers

    Penglin Dai and co-authors describe AutoMCU, an LLM multi-agent system that designs neural networks to fit a microcontroller's RAM and flash limits. Vendor toolchain checks reject candidates that would not fit before any training. The paper reports competitive CIFAR accuracy after 1 to 2 hours of customization and tests on several STM32 parts; no code is linked.